(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Proceedings of the Society for Experimental Biology and Medicine"

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 



at |http : //books . google . com/ 



4 
A 









-<.-'* 

'>•.; 



'^'IZilP^^- 






ISSN 0057-9727 



Volume 171, Number I, October 1982 



PROCEEDINGS OF 
THE SOCIETY FOR 

Experimental 
Biology and 
Medicine 




ACADEMIC PRESS 



New York Lofidan 
PlUns SftJI DicdLLi SjH FiiiJliLiscu !?aci Pituttt SvJnicv TiiKvu Ttin>nir 



The Society was founded by S. J. Meftzer in t9()3 



Council 



President, Theodore Cooper '83 
The Upjohn Company 

President-Elect, Marjorie B. Zucker '83 
New York University 

Past-President, Robert W. Berliner '83 
Yale University 

Treasurer, Harriet B. Dustan '84 
University of Alabama 

Executive Secretary, Mero R. Nocenti (ex officio) 
Columbia University 



Lloyd L. Anderson '83 
Iowa State Univ. 

R. AUERBACH '83 

Univ. of Wisconsin 

H. A. Bern '83 
Univ. of California 

Walter J. Bo '85 
Bowman-Gray Sch. of Med. 

M. G. Horning '83 
Baylor Coll. of Med. 



John H. Laragh '86 
Cornell Med. Ctr. 

Julian B. Marsh '85 
Med. Coll. Pennsylvania 

S. M. McCann '83 
Univ. of Texas 

John Resko '85 
Univ. of Oregon 

Wilbur H. Sawyer '86 
Columbia University 



Arthur A. Spector '86 
Univ. of Iowa 

J. Tepperman '83 

State Univ. of New York 

A. K. Weiss '83 
Univ. of Oklahoma 
Chairman, Membership Comm. 

M. B. Zucker '82 
New York Univ. 
Chairman, Publ. Comm. 



membership application 

Membership in the Society for Experimental Biology and Medicine is open to all individuals who 
have independently published original meritorious investigations in experimental biology or experi- 
mental medicine and who are actively engaged in experimental research. In general, applicants should 
be beyond a supervised postdoctoral experience in order to be able to demonstrate the ability to 
conduct independent investigations. 

Associate Membership is available for individuals who are engaged in research in experimental 
biology or medicine or in teaching or administration in these areas. Student Membership is available to 
any student who is a candidate for a doctorate degree and is actively engaged in research in experi- 
mental biology or medicine. Write to the Executive Secretary of the Society for details about eligibility 
requirements and application procedures for Associate and Student Memberhips. 

Application forms may be obtained from the Office of the Executive Secretary, Society for Experi- 
mental Biology and Medicine, 630 W. 168th St., N.Y., N.Y. 10032. In addition, a tear-out applica- 
tion form is printed in each issue of the PSEBM. 



Proceedings of The Society for Experimental Biology and Medicine. Vol. 171, No. 1. October 1982. Published 
monthly except August by Academic Press, Inc., Ill Fifth Avenue. New York. N.Y. 10003. Second class postage 
paid at New York. N.Y. and at additional mailing offices. 1982: Subscription per year $68.00 U.S.A.; $88.75 out- 
side U.S.A. 1983: Subscription per year $73.00 U.S.A. and Canada; $%.00 outside U.S.A. and Canada. All prices 
include postage and handling. Send notices of change of address to the Office of the Publisher at least 6-8 
weeks in advance. Please include both the old and new addresses. Postmaster: Send all changes of address to 
Proceedings of the Society for Experimental Biology and Medicine, 111 Fifth Avenue, New York, N.Y. 10003. 
Copyright © 1982 by the Society for Experimental Biology and Medicine. 



^^'^^ 



R. P. Ahlquist 
J. C. Allen 
F. Allison, Jr. 
E. Alpen 
R. L. Baehner 

C. A. Barraclough 
A. L. Barron 

R. A. Berg 

D. BOGGS 
Y. BOREL 

K. G. Brand 

E. Bransome 

E. Bresnick 
S. M. Cain 
H. F. Clark 
I. Clark 

K. H. Clifton 
M. L. Cohen 
C. W. Cooper 

R. A. CORRADINO 

R. J. Cousins 
N. E. Cremer 

F. A. DeLustro 
T. C. Detwiler 
J. DiSalvo 

S. Ebbe 

G. Eknoyan 

J. D. Finkelstein 
J. W. Fisher 

P. P. FOA 

E. L. Forker 
H. M. Frankel 
R. R. Gala 
R. C. Gallo 



Board of Editors 



Editor 

M. R. NOCENTI 

630 W. 168th Street 

New York, N.Y. 10032 

212 795-9223 

Associate Editor 
Bernard F. Erlanger 

Columbia University 



Z. N. Gaut 

A. L. Goldstein 

E. C. GOTSCHLICH 

S. Greenberg 

M. R. C. Greenwood 

C. E. Grosvenor 

G. GUROFF 

N. S. Halmi 
M. A. Heymann 

M. R. HlLLEMAN 
S. T. HOFFSTEIN 

J. J. Holland 

J. A. HOLOWCZAK 

J. B. Hook 

D. HORROBIN 
K. HUBER 

E. D. Jacobson 
H. D. Janowitz 
D. C. Johnson 
R. C. Johnson 
T. J. Kindt 

F. G. Knox 

C. A. Krakower 
L. C. Krey 

M. KUSCHNER 

P. L. LaCelle 
C. A. Lang 
C. Lenfant 
R. Levere 
J. V. Levy 
J. Litwack 
P. D. Lotlikar 

G. J. Macdonald 
1. Mandl 

J. J. Marchalonis 



A. J. Marcus 
R. E. McCaa 
S. M. McCann 
J. C. McGiFF 
G. Medoff 

J. Meites 

T. C. Merigan, Jr. 

F. N. Miller 

C. R. Morgan 

P. J. MULROW 

L. H. Muschel 

G. D. NiSWENDER 

B. L. O'Dell 
S. Oparil 

R. OSTWALD 

p. Y. Paterson 

V. A. Pedrini 

J. C. Penhos 

J. M. Phillips-Quagliata 

G. L. Plaa 

S. A. Plotkin 

D. D. Porter 

A. S. Rabson 
J. A. Ramaley 
R. J. Reiter 

J. A. Resko 
J. A. Rillema 
R. B. Roberts 
R. Ross 
J. Rudick 
T. M. Saba 
W. Sawyer 

B. B. Saxena 
A. J. Sbarra 
A. V. Schally 



R. A. SCHEMMEL 
R. SCHMID 

N. J. Schmidt 

C. A. Schneyer 
W. N. Scott 

J. E. Sealey 

E. E. Selkurt 

L. Sensenbrenner 

J. H. Shaw 

E. M. Shevach 

N. Shock 

M. M. Sigel 

G. W. SiSKIND 

J. L. Spivak 

M. B. Stemerman 

J. G. Stevens 

B. J. Stover 

G. P. Studzinski 
G. A. Tanner 
G. J. Thorbecke 
M. L. Tyan 
J. L. Vaitukaitis 

D. R. Van Campen 
S. F. Vatner 

C. M. Veneziale 
R. L. VicK 

S. R. Wagle 
M. E. Weksler 
J. M. Weller 
R. M. Welsh, Jr. 

D. L. Wiegman 

D. B. ZiLVERSMIT 

B. G. Zimmerman 

M. B. ZUCKER 



Publication Committee 



M. ZucKER, '82, Chairman, M. Hilleman, '82; 

S. NiEWiARowsKi, '84; S. Seifter, '82; and J. Tepperman, '86. 

-ru^ President, President-Elect and Executive Secretary 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL 
BIOLOGY AND MEDICINE 

Volume 171, Number 1, October 1982 

Copyright © 1982 by the Society for Experimental Biology and Medicine 
All Rights Reserved 

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or 
mechanical, including photocopy, recording, or any information storage and retrieval system, without permis- 
sion in writing from the copyright owner. 

The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright 
owner*s consent that copies of the article may be made for personal or internal use, or for the personal or internal 
use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy 
fee through the Copyright Clearance Center, Inc. (21 Congress Street, Salem, Massachusetts 01970) for copying 
beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other 
kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating 
new collective works, or for resale. Copy fees for pre- 1982 articles are the same as those shown for current 
articles. 

MADE IN THE UNrTED STATES OF AMERICA 

Send requests for permission to reproduce items published in Proceedings of the Society for Experimental Biol- 
ogy and Medicine to: Dr. Mero R. Nocenti, Editor, Society for Experimental Biology and Medicine, 630 W. 168th 
St., New York, NY 10032. 



INDEX TO ADVERTISERS 

Members and subscribers are requested to cooperate with our advertisers 
Charies River Cover 4 



TABLE OF CONTENTS 
MINIREVIEW 

Concerning the Study of Nutrition in China H. C. Hou 1 

ENDOCRINOLOGY 

Effects of Culture Age on PRL and GH Responses to K. Hanew, E. G. Rennels 12 

Bromocriptine and Somatostatin from Primary 
Cultures of Rat Anterior Pituitary Cells 

GROWTH AND DEVELOPMENT 

Differences in the Reported Frequencies of Cleft Lip plus M. L. Tyan 41 

Cleft Lip and Palate in Asians Bom in Hawaii and the 
Continental United States 

HEMATOLOGY 

Phosphoenolpyruvate (PEP) Effects on Fresh and Stored P. R. Sohmer, R. L. Scott 24 

Red Blood Cells 
Cobalamin (Vitamin Bjj) Analogs Are Absent in Plasma J. van der Westhuyzen, F. Fernan- 

of Fruit Bats Exposed to Nitrous Oxide dez-Costa, J. Metz, G. Drivas, V. 

Herbert 88 

MICROBIOLOGY/IMMUNOLOGY 

Cold Agglutinin Antibody in a Hyperimmune Erythrocyte K. D. Warber, J. C. Brown 7 

Antiserum 

Light and Electron Microscopic Studies of the Pathogenesis D. C. Bosse, W. G. Campbell, Jr., 

of Vaccinia Virus Infection in Mouse Brain W. A. Cassel 72 

Concomitant Enhancement of B-Cell Mitogenesis and T. A. Ferguson, L. A. Fish, C. S. 

Inhibition of Antibody Synthesis by a Phorbol Ester Baxter, J. G. Michael 83 

Therapeutic Concentrations of Antineoplastic Agents T. C. Cesario, L. M. Slater, H. S. 

Diminish Interferon Yields Kaplan, J. G. Tilles 92 

Immune Response to Laminin, a NoncoUagenous Glyco- A. M. Mackel, F. DeLustro, E. C. 

protein of Basement Membrane, in a Syngeneic LeRoy 98 

Murine System 

Spontaneous Production of High Levels of Leukocyte D. V. Ablashi, S. Baron, G. Arm- 

(a) Interferon by a Human Lymphoblastoid B-Cell strong, A. Faggioni, D. Viza, 

Line (LDV/7) P. H. Levine. G. Pizza 114 

ONCOLOGY 

Localization of Radiolabeled Antibody in SVT2 Tumor D. J. Buchsbaum, J. M. Anderson, 

Increases with Immunosuppression of the Host B. E. Bray 56 

PATHOLOGY 

The Yucatan Miniature Swine: An Improved Pig Model J. M. Terris, R. C. Simmonds 79 

for the Study of Desoxycorticosterone-Acetate 
(DOCA) and Aldosterone Hypertension 

PHARMACOLOGY 

Immunocytochemical Localization of Catechol-O- Methyl- K. Inoue, C. R. Crevling, L. W. Tice 65 
transferase in Rat Parotid Gland 

PHYSIOLOGY 

The Transport of Urate in the Small Intestine of the Rat C. E. Dukes, D. A. Steplock, A. M. 

Kahn, E. J. Weinman 19 



Effects of Sodium Chloride on Prostacyclin-Stimulated D. Villarreal. R. H. Freeman, J. O. 

Renin Release in Dogs with Filtering and Nonfiltering Davis, J. R. Dietz, S. F. Ech- 

Kidneys tenkamp 34 

Ovine Maternal and Fetal Circulatory Responses to an D. B. Schwartz, T. M. Phernetton, 

Endoperoxide Analog M. K. Stock, J. H. G. Rankin 46 

The Sex-Related Difference in Perfluorooctanoate Excre- H. HanhijArvi, R. H. Ophaug, 

tion in the Rat L. Singer 50 

Exercise and Estrous Cycle Influences on the Plasma W. K. Palmer, J. R. Davis 120 

Triglycerides of Female Rats 

TISSUE CULTURE 

Factors Modifying DNA Synthesis by Lung Fibroblasts C. W. Castor, T. D. Fremuth 109 

in Vitro 



NORTH EMGINEE.^ING LIBRARY 

1002 I.S.T. BLDG. 
THE UNIVERSITY OF MICHtGAN 
ANN ARBOR, MICHIGAN 48109 



STUDY OF NUTRITION IN CHINA 



published mainly in such journals as the 
Chinese Journal of Physiology, the Chinese 
Medical Journal, Journal of the Chinese 
Chemical Society, Nutrition Bulletin of 
Peking Agriculture Institute, Bulletin of the 
Institute of Biology of the Sino Academia, 
etc. In 1926 nutrition workers participated 
in the organization of the Chinese Physio- 
logical Society, which was inaugurated in 
February of the same year; thus it was 
founded about 2 years prior to the inaugu- 
ration of the American Institute of Nutri- 
tion. But it was not until December 1946 
that nutritionists of the Chinese Physiologi- 
cal Society inaugurated the first Chinese 
Nutrition Society and subsequently pub- 
lished the Chinese Journal of Nutrition. 
Prior to that time, the Chinese Journal of 
Physiology published many articles on nu- 
trition. 

The years from 1938 to 1950 in the history 
of Chinese nutritional work might be 
termed the period of hard struggle, when in 
the first 7 years our country was invaded by 
the Japanese. Universities, colleges, and 
research institutes were moved to the 
interior and nutritional investigation was 
continued under very difficult conditions. 
In spite of the lack of adequate facilities 
such as laboratory space and equipment, 
nutritional work was carried on in tempo- 
rary quarters. Prevalence of nutritional de- 
ficiency diseases due to the lack of proper 
foods prompted greater effort in nutritional 
studies. Two national nutrition conferences 
were called to discuss the basic nutritional 
problems of the country in 1941 and 1945, 
with the presentation of a number of nutri- 
tional research papers and the formal adop- 
tion of the first Chinese dietary nutrition 
requirements that were written by the Nu- 
trition Committee of the Chinese Medical 
Association in 1938 (4). 

After the surrender of the Japanese in- 
vaders in 1945, universities and research in- 
stitutes were moved back to their original 
sites. Even though nutritional problems had 
become more acute because of the politi- 
cally unsettled conditions and the difficult 
economic situation, urgently needed nutri- 
tional studies did not progress as they 
should have. The nutritional status of the 



great mass of people remained deplorable. 
Soon after the overthrow of the old re- 
gime and the establishment of the People's 
Republic of China, great attention was paid 
by the government to the nutritional prob- 
lems of the people. The nutritional status of 
the people improved considerably. Nutri- 
tional deficiency diseases became rare in 
cities, though in some remote rural areas, 
owing to ignorance and improper intake of 
foods, some deficiency diseases such as 
pellagra, infantile beriberi, goiter and 
xerophthalmia were still present. Studies in 
basic and applied nutrition were encour- 
aged and promoted by the new government, 
and much was accomplished. In April of 
1954, the Chinese Nutrition Society merged 
with the Physiological, Biochemical, Phar- 
macological, and Biophysical Societies to 
form the Chinese Society of Physiological 
Sciences and held local monthly as well as 
national annual scientific meetings. In 1956 
the reorganized Chinese Journal of Nutri- 
tion appeared as a quarterly. Theoretical, 
basic, and applied nutrition progressed 
rapidly until the cultural revolution was 
usurped by Lin Biao and the ''Gang of 
Four," and nutritional work was much de- 
precated. Many nutritional workers were 
forced to suspend activities or to undertake 
other jobs. We actually lost over 10 years of 
otherwise very productive work. Nutri- 
tional Society activities, including the jour- 
nal publication, were forced to suspend op- 
erations. After the dc*wnfall of the ''Gang of 
Four,'' nutritional work was gradually re- 
vived, and research resumed. In October of 
1978, the Chinese Society of Physiological 
Sciences held a one-week conference, in 
which papers on "recent advances in nutri- 
tion,'' on "diet and nutrition during the 
'New Grand March,' " and on "nutritional 
education and research" were presented 
and discussed. Over 70 research papers 
were submitted to the nutritional section of 
the conference. In September of 1979 a na- 
tional nutrition conference was held in 
Cheng Du in which important problems of 
nutrition, including dietary nutritional re- 
quirements and allowances, new dietary 
protein and lipid sources and nutrition for 
children and the aged were discussed. In 



STUDY OF NUTRITION IN CHINA 



addition, there were over 180 research pa- 
pers on subjects relating to basic nutrition, 
childhood nutrition, nutrition and diseases, 
new sources of foods, the effect of pro- 
cessing and preservation on food values, 
nutritional survey and food hygiene sub- 
mitted to the conference. In October of 
1979 a national conference on nutritional 
requirements and dietary allowances was 
convened in Tian Jin, and a tentative, re- 
vised list of nutritional allowances was for- 
mulated and discussed. The final form of 
the revised recommended dietary allow- 
ances was approved at the Third National 
Nutritional Scientific Conference held in 
May 1981, and topics of special interest 
were also discussed, research papers (over 
300 were submitted) were presented, and 
the National Nutrition Society was formally 
inaugurated. Thus nutritional work has 
been reinvigorated, and it is hoped that be- 
fore long nutrition activities will be ex- 
panded greatly to meet the actual need of 
the whole country. 

Past Nutrition Studies. The first publi- 
cation on nutrition investigation in China 
was "Dietary Studies in Shantung'' by 
Adolph in 1913 (5). This was followed by a 
paper on the analysis of litchi by Read (6) in 
1918 and a paper on a study of the different 
kinds of milk in Kwangtung in the same 
year (7). Papers on the nutritive values of 
soybean products and analyses of Chinese 
foods appeared in 1920 (8) and 1921 (9). 
During subsequent years, many more pa- 
pers on diet and nutrition in China were 
published on varied subjects: (i) dietary 
studies in different parts of China, in urban 
and rural areas, among different groups of 
people (students, institution workers, fac- 
tory workers, farmers, soldiers and labor- 
ers) were undertaken; (ii) analyses on pro- 
tein, fat, carbohydrate, mineral, vitamin, 
water, and crude fiber content of foods in 
different parts of China were done. Consid- 
erable variation in composition of foods 
from different localities, especially vitamins 
and certain minerals, were found. Owing to 
the vastness of our country, great differ- 
ences in climatic conditions, soil composi- 
tion, use of fertilizers, selection of genetic 
strains, agricultural methods, etc., such 



variations would be expected. Other topics 
of research were: (iii) digestion, utilization, 
and metabolism among different groups of 
nhabitants and in healthy and diseased in- 
lividuals; (iv) protein and amino acids in 
nutrition; (v) vitamins in foods and nutri- 
tion; (vi) minerals in foods and nutrition; 
(vii) nutrition in vulnerable population 
groups, particularly children; (viii) nutri- 
tional value of vegetarian diets; (ix) dietary 
requirements and allowances for the Chi- 
nese; (x) nutritional deficiency diseases 
among Chinese under different circum- 
stances such as peacetime, wartime, fam- 
ine years, political and economic distur- 
bances, particularly deficiencies in vita- 
mins A, D, Bi, C, and nicotinic acid as well 
as protein and iodine deficiencies. Patients 
with deficiencies were found to be much 
more numerous during difficult times than 
during peaceful times. Introduction of 
iodized salt or other iodine-containing 
foods greatly reduced the incidence of goi- 
ter in endemic areas. In addition to the 
finding of iodine-deficiency goiter, goi- 
terogenic substances were found in cer- 
tain foods (10). 

Aside from the above nutritional studies 
there were also organizations or units for 
the promotion of applied nutrition, nutri- 
tional propaganda, and education, with 
provision for free distribution of food at 
times of emergency and supplementary 
feeding of the poor to combat deficiency 
diseases (11, 12). 

Following the establishment of the Peo- 
ple's Republic of China in 1949, consid- 
erable work on basic nutritional research, 
applied nutrition, and dietary therapy was 
conducted at various institutions. The 
achievements or results of studies were 
summarized in several papers published in 
the 10th anniversary publication of the 
Ministry of Health (13) and various science 
journals such as the Progress in Physiologi- 
cal Sciences (14), the Chinese Medical 
Journal, and the Journal of the Chinese 
Chemical Society, In brief, advancement was 
attained in investigations on dietary and nu- 
trition surveys, on dietary and nutrition re- 
quirements and allowances, that formed the 
basis for the first revision of Chinese di- 



STUDY OF NUTRITION IN CHINA 



etary allowances; nutritional deficiency 
diseases, their treatment and prevention; 
dietary therapy and special diets for pa- 
tients with various diseases; food composi- 
tion analyses including amino acids, fatty 
acids, vitamins, and toxic substances; im- 
provement of nutritional survey tech- 
niques; influence of cooking, storage, pro- 
cessing, and preservatives on the nutritive 
values of foods; effects of environment on 
nutrition, etc. It was found, for instance, 
that certain algae were suitable as a sup- 
plementary food (15), that low doses of 
gamma ray radiation for preservation of 
foods (grains) did not influence the nutritive 
values nor impart any toxicity to the ir- 
radiated foods (16), and that the nicotinic 
acid in com was found to exist in bound 
form and could be liberated as the free form 
by the simple addition of sodium bicarbo- 
nate during cooking and thus become ef- 
fective in preventing pellagra (17). Keshan 
disease, an endemic chronic myocardiopathy 
was found to be related mainly to a lack of 
sufficient selenium (18). The soil and foods 
of the endemic areas as well as the blood 
and hair of the inhabitants were found to be 
very low in selenium content, and oral in- 
take of small amount of selenium salt 
greatly lowered the incidence of Keshan 
disease (19). 

Current Nutritional Studies. Both 
basic (theoretical) and applied nutritional 
studies are being carried out at institutions 
throughout the country on such general 
topics as: 

1. The relation between diet, nutrition, 
and diseases such as studies on foods low- 
ering blood lipids and their relation to coro- 
nary heart diseases; on the use of amino 
acid mixtures or fish protein concentrate in 
patients with injuries and in malnutrition 
cases; further studies on the use of the hy- 
brid com from opaque-2 with a Chinese va- 
riety for the prevention and treatment of 
pellagra; on the cause of 'hyperiodine goi- 
ter/' etc. 

2. Food fortification with stable vita- 
mins, amino acids, calcium, and iron in 
health and disease. 

3. Food hygiene, food contamination. 



food toxicology, insecticide residues in re- 
lation to nutrition. 

4. Further studies on food compositions 
in newer foods, especially with regard to 
amino acids, fatty acids, vitamins, and 
trace elements. 

5. The interrelationship between nutri- 
ents in diets of different compositions in 
different ethnic groups throughout the 
country. 

6. Nutritional requirements and dietary 
allowances of various age and sex groups 
among different ethnic groups. 

7. Nutritive values of newer vegetable 
proteins and novel foods. 

8. The relation of diet and nutrition to 
cancers. 

9. Further studies on nutritional factors 
in Keshan disease. 

10. Nutritional problems in relation to 
agriculture and animal husbandry. 

In addition to the above topics, there are 
certain problems that deserve further con- 
sideration: 

1. The art of Chinese cooking is well 
known throughout the world as being tasty, 
colorful, and artistic, but the influence of 
the mode of processing and cooking on nu- 
tritive values has not yet been well studied. 

There are many old methods of preserv- 
ing foods at times of abundance to meet the 
need at times of want. Preservation alters 
the taste and appearance of foods. So far 
changes in nutritive values during preser- 
vation have not been thoroughly investi- 
gated. 

It will be very interesting and of practical 
importance to make a thorough study of the 
various methods of preparing and preserv- 
ing foods in China with regard to any possi- 
ble alteration in nutritive value. 

2. The incidence of certain cancers is 
particularly high in certain parts of China. 
There is every indication that it is related to 
food habits, the prolonged intake of harmful 
ingredients in certain foods or the continual 
lack of certain protective nutritive sub- 
stances in diets. Further studies may yield 
fruitful results. 

3. Recent surveys indicate that goiter 
and cretinism are still present in certain 



STUDY OF NUTRITION IN CHINA 



parts of China. This is due mainly to a lack 
of proper enforcement of the use of iodized 
salt in endemic areas; it had been the prac- 
tice in certain districts to roast salt before 
use, thus causing loss of the added iodine. 
It has been noted recently that in certain 
oil field areas, deep well water containing a 
very high level of iodine can also induce 
goiter. The cause of this needs further 
study. The possibility of the presence of 
other goiterogenic agents has yet to be 
ruled out. 

4. The lack of selenium has been impli- 
cated in Keshan disease, but the presence 
of other factors, nutritional or otherwise, 
might play a part. Progress has been made 
in the elucidation of the multiple factors in 
its etiology but further clarification is still 
needed. 

5. The last country-wide nutritional sur- 
vey was made over 20 years ago. The re- 
sults of that survey were not well assessed. 
The present nutritional status of the whole 
country is yet unknown. A country-wide 
survey in 1982 is planned, the goal of which 
will be to obtain a concrete idea of the ac- 
tual nutritional status to be used as a basis 
for the planning of food production and 
distribution. 

6. Owing to the differences in climatic 
conditions, soil qualities, fertilizers used, 
and methods of cultivation, the nutrient 
composition of foods of different localities 
varies considerably. Thus the present food 
composition tables cannot be used in differ- 
ent districts throughout the country. Many 
more food-analysis data are required, espe- 
cially with regard to trace elements, amino 
acids, fatty acids, and vitamins. 

7. There are differences in the dietary 
habits and ways of livelihood of different 
tribes of people in different localities of the 
country, and the dietary requirements 
likely vary. Many more studies are thus 
needed in order to establish a better under- 
standing of the real nutritional need of 
every individual and the population as a 
whole. 

8. The production of food in our country 
at present does not keep up with the growth 
of the population (birth control is already in 



practice). It is necessary therefore to search 
for newer foods (especially protein-rich 
food sources), to conserve food (avoid 
waste), and to use the right combination of 
foods in order to obtain the highest degree 
of digestion and utilization so that no food 
is wasted. Obesity due to overeating and 
certain associated diseases are not uncom- 
mon now. Hence much more nutritional 
education is needed, both in the training of 
nutritional workers and in the dissemina- 
tion of nutritional knowledge to the masses. 
There is also a need for closer cooperation 
between nutritional workers and food pro- 
ducers (agriculturists, food manufacturers, 
and animal husbandry workers). 

1. Huang Di Nei Jin Su Wen, revised by Kao Bao 
Heng et al. People's Health Publishing House, 
Vol 7(8):pl49, 1%3. 

2. Hou HC. Chinese ancient books on food and nu- 
trition. Nat Med J China 22(11): 1015- 1026, 1936. 

3. Hou HC. Dietary principles in ancient Chinese 
medicine. Chin Med J 57:347-352, 1938; Beriberi 
in ancient Chinese medical literature. Chin Med J 
58:302-313, 1940. 

4. Committee on Nutrition. Chinese Medical As- 
sociation Special Report, Ser. No. 10. Shanghai, 
1938. 

5. Adolph WH. Diet studies in Shangtung. China 
Med J 27:1013, 1913. 

6. Read BE. Analysis of Litchi. J Amer Chem Soc 
40:817, 1918. 

7. Levine CO, et al. A study of the different kinds of 
milk production in Kwangtung. China Med J 
32:536, 1918. 

8. Wilson SD. A study of Chinese foods. China Med 
J 34:503, 1920. 

9. Embrey A, Wang TC. Analysis of some Chinese 
foods. China Med J 35:247, 1921. 

10. Hou HC. Thyroid enlargement following liver 
feeding in rats. Proc Soc Exp Biol Med 43:753, 
1940. 

11. Hou HC. Chinese dietary standard and some di- 
etary problems among war refugees. C R Congr 
FEATM, 10th, Hanoi, November 26- December 
2, 1938. 

12. Hou HC. Nutrition supplements for refugee chil- 
dren. Chin Health Mag 2(4), 1940. 

13. Collected Papers of Medical Researches in Cele- 
brating the 10th Anniversary of the Founding of 
New China. Peking, Ministry of Health of the 
People's Republic of China, Vol 1, Decem- 
ber 1959. 



STUDY OF NUTRITION IN CHINA 



14. Hou HC. Ten years achievement in vitamin re- 
search. Prog Physiol Sci 3(2):105-115, 1959. 

15. Hou HC, et al. Studies on the nutritional values 
of Chinese algae. Chin Health J 8(1): 29-34, 1963. 

16. Hou HC, et al. On the question of toxicity of 
gamma ray irradiation upon grains. Technique in 
nuclear energy — Applied radiation with isotope. 
Ppl67-173, 1960. 

17. Wang GY et al. Studies on improving the utiliza- 
tion of nicotinic acid in corn. Chin J Physiol 
29:197, 1966. 

18. Keshan Disease Research Group, Chinese Acad- 
emy of Medical Sciences. Studies on the epi- 



demiological characteristics on the etiological 
relationship of selenium to Keshan disease. Nat 
MedJChina59(8):451, 1981. 

19. Research Laboratory of Keshan Disease Group, 
Xian Medical College, et al. Observation on the 
effects of sodium selenite for preventing acute 
Keshan disease. Nat Med J China 59(8):457, 1979. 

20. Zhu XY. Development of ongoing research work 
on endemic goiter and cretinism. Nat Med J China 
6O(12):705-707, 1980. 

Received February 5, 1982. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171. 7-11 (1W2) 

Cold Agglutinin Antibody in a Hyperimmune Erythrocyte Antiserum (41469) 
KIMBROUGH D. WARBER and JOHN CLIFFORD BROWN* - 

Department of Microhioioay, University of Kansas, Lawrence, Kansas 66045 



Abstract. To examine whether an innocuous, nontvicterial-associated immunogen could 
generate a cold agglutinin autoreactive antibody response, rabbits were hyperimmunized 
with sheep erythrocytes. Of five animals immunized, one developed an autoreactive 
antibody population which agglutinated autologous and allogenic erythrocytes at tempera- 
tures below 37°. The cold agglutinin antibody activity was present in the IgM class immuno- 
globulin fraction since only the excluded volume of a Sephacryl S-200 serum fractionation, 
and purified IgM, contained detectable activity. The cold agglutinin antibody was hemolytic 
in the presence of guinea pig complement, since whole serum, and purified IgM, sensitized 
rabbit erythrocytes for lysis in a biphasic temperature hemolysis assay. In hemolytic inhibit 
tion assays, the cold agglutinin antibody was sugar specific. The relative sugar ligand speci- 
ficity in this assay was shown to be ^-acetylgalactosamine "^ melibiose "^ galactose "^ 
lactose. The hyperimmune anti-sheep erythrocyte serum agglutinated both sheep erythro- 
cytes and Group C streptococcal vaccine at 4** and at 37°. These data suggest certain an- 
imals can respond to nonbacterial-associated immunogen with antibody specific for the im- 
munogen, but which may cross-react with autologous sugar determinants. 



Cold agglutinin (autoreactive erythro- 
cyte) antibodies have been shown to be as- 
sociated with infectious illness and other 
human disease (1-5), and may appear 
spontaneously in certain strains of mice (6). 
Also, cold agglutinin antibodies may appear 
during hyperimmunization of rabbits with 
certain bacterial vaccines (7-9). Because 
of these associations, it has been of interest 
to determine whether these particular au- 
toantibodies arise in response to the infec- 
tious or immunizing agent or whether they 
arise nonspecifically as one result of gen- 
eral immune stimulation. 

A recent investigation by our laboratory 
revealed a cold agglutinin antibody re- 
sponse in rabbits hyperimmunized with 
Group C streptococcal vaccine (10). The 
Group C streptococcal cell wall carbohy- 
drate structure carries determinants identi- 
cal to those present on the glycan portion of 
the Forssman glycolipid of sheep erythro- 



* This work was supported by U.S. Public Health 
Service Grant AI 16220 from the National Institutes of 
Health. 

* J.C.B. is the recipient of Research Career Devel- 
opment Award 1-K04-AI00427 from the National In- 
stitutes of Health. 



cytes. These determinants are a-anomer- 
ically linked ^-acetylgalactosamine di- 
saccharide residues. These same residues 
are the immunodominant structures present 
on Group C carbohydrate (11). Both IgM 
and IgG cold agglutinin antibodies isolated 
from Group C streptococcal antisera re- 
acted with ^-acetylgalactosamine and 
Group C carbohydrate. These data there- 
fore suggested that these particular cold 
agglutinin antibodies were induced in re- 
sponse to the Group C streptococcal cell 
wall carbohydrate. 

The present investigation was initiated to 
examine whether a bacterial vaccine was 
necessary to induce the cold agglutinin re- 
sponse, or whether a similar antibody re- 
sponse might be generated by hyperim- 
munization of rabbits with sheep erythro- 
cytes. 

Materials and Methods. Vaccine prepa- 
ration and immunization protocol. Prepa- 
ration of vaccine was according to Lance- 
field (12). Briefly, bacteria were cultured 
overnight in Todd- Hewitt broth and vac- 
cine prepared from washed, pepsin-di- 
gested cells. The procedure used for im- 
munization was according to Herd and 
Spragg (13). Rabbits were immunized three 



0037-9727/82/090007.05$0 1 .OCWQ 
AW hght« TeMTveA. 



8 



COLD-REACTIVE ANTIBODY 



times per week for 4-5 weeks, rested 3 
months, and again immunized. 

Sheep erythrocyte immunization pro- 
tocol. Sheep erythrocytes in modified 
Alsever's solution were washed three times 
in 0.15 A/ NaCl and pelleted by centrifuga- 
tion at 800 g. Red blood cells were resus- 
pended in 0.15 M NaCl to 10% (v/v) and 1.0 
ml injected iv three times per week for 5 
weeks into five animals. Rabbits were bled 
from the central ear artery each week. 

Hemagglutination assay. Allogenic rab- 
bit erythrocytes were collected in Alsever's 
solution and washed in 0.02 M phosphate, 
0.15 M NaCl buffer, pH 7.2, made 1% (w/v) 
in bovine serum albumin. Cells were resus- 
pended in this buffer to 0.5% (v/v) and dis- 
tributed in * W type microtiter plate wells. 
The wells contained an equal volume of two- 
fold-diluted whole or Sephacryl S-200 frac- 
tionated sheep erythrocyte antisera. Mi- 
crotiter trays were stored at 4° overnight 
and the reciprocal of the highest dilution 
which yielded positive agglutination was 
recorded as the cold agglutinin antibody 
titer. 

Biphasic temperature hemolytic assay. 
This assay has previously been described in 
detail (10). Essentially, a pretitrated excess 
amount of guinea pig complement was 
added to a 0.25% (v/v) suspension of rabbit 
erythrocytes in the presence of twofold se- 
rially diluted whole or fractionated antisera. 
After incubation at 4° for 60 min and 37° for 
60 min the amount of hemolysis was deter- 
mined by supernatant volume absorbance 
measurements at 413 nm. Inhibition in this 
assay was performed using a pretitrated 
amount of cold agglutinin antibody suffi- 
cient to yield 50% hemolysis under con- 
ditions described above. Preincubation 
of increasing amounts of varying inhibitors 
with this particular amount of antibody 
was performed for 3 hr at 4° prior to assay. 
Inhibition was calculated relative to an un- 
inhibited control. 

Polysaccharide preparations. Group C 
streptococcal carbohydrate was purified 
according to an acidified sodium nitrite 
procedure provided by E. Gotchlisch, The 
Rockefeller University, New York (per- 
sonal communication). Type III pneu- 



mococcal polysaccharide was purified 
according to Campbell and Pappenheimer 
(14). Type XIV pneumococcal polysac- 
charide was the generous gift of Eli Lilly 
and Company, Indianapolis, Indiana. 

Results. Rabbit erythrocyte agglutinating 
activity at 4** or 37° was absent from pre- 
inoculation sera, while at these temper- 
atures the same sera agglutinated sheep 
erythrocytes to a 1:4 dilution. After im- 
munization, all rabbits produced sheep red 
blood cell-reactive antibodies (reactive at 
both 37** and 4°). Only one rabbit, however, 
produced antibodies reactive against rabbit 
erythrocytes; and, reactivity was detect- 
able only at 4°. For this particular rabbit's 
antiserum, the sheep erythrocyte hemag- 
glutination titer increased from 1:4 to 1:1024 
during the fifth week of immunization. 
In contrast, hemagglutination reactivity 
of whole serum with rabbit erythrocytes 
occurred during the first, second, and third 
weeks only, and was relatively weak (high- 
est titer < 1:32), and occurred only at 4**. 
Reactivity with both human adult O- 
positive and cord blood erythrocytes oc- 
curred in the warm (titer < 1:4) and in- 
creased in the cold (titer < 1 :8). None of the 
remaining animals' sera reacted with 
human or rabbit erythrocytes when tested 
at either 37° or 4°. Hemolytic activity to- 
ward rabbit erythrocytes in the biphasic 
temperature hemolytic assay reached, 
however, a titer of 1:256 during the third 
week of immunization. Since most cold 
agglutinin antibodies are of the IgM class, 
the anti-sheep red cell antiserum was frac- 
tionated into 19 S (excluded) and 7 S (in- 
cluded) components by Sephacryl S-200 
chromatography and examined for erythro- 
cyte autoreactivity. 

Figure 1 shows the results of hemolytic 
assays performed upon individual Seph- 
acryl S-200 fractions. As shown, it is clear 
that all of the cold agglutinin activity was 
confined to the excluded column fraction; 
these results in addition to Ouchterlony 
analysis with goat anti-rabbit /ll chain anti- 
serum and polyvalent goat anti-rabbit Ig 
antisera, showed that IgM was the only 
immunoglobulin component in these frac- 
tions. That the cold agglutinin activity was 



COLD-REACTIVE ANTIBODY 




30405060 70 8090 
Fraction Number 

Fig. 1 . Fractionation of 4.0 ml rabbit sheep eryth- 
rocyte antiserum on a 2.8 x 100-cm Sephacryl S-200 
column. Shaded area represents fractions which ex- 
hibited cold agglutinin activity. Fraction volumes were 
4.1 ml. 



associated with sheep erythrocyte determi- 
nants was shown by absorption experi- 
ments. Whole antiserum was absorbed once 
in the presence of one-half the serum vol- 
ume of packed erythrocytes. In addition, 
the same procedure was performed using 
Group C streptococcal vaccine. After ab- 
sorption, each solution was titrated in the 
biphasic temperature hemolysis assay. 
Shown in Fig. 2 are the results of these as- 
says. As depicted, both absorption proce- 
dures were effective in depleting the anti- 
sera of hemolytic activity toward rabbit red 
blood cells. Absorption with Type III 
pneumococcal vaccine did not alter the 
original activity. Because of the apparent 
reactivity of the cold agglutinin material 
with sheep erythrocytes and Group C 
streptococcal vaccine, further examination 
involved fractionated material in ligand in- 
hibition experiments. Cold agglutinin an- 
tibody, after fractionation by Sephacryl 
S-200 chromatography, was concentrated 
and the IgM antibody titrated to yield 50% 
hemolysis in the biphasic temperature he- 
molysis assay. This amount of antibody 
was subsequently preincubated in the pres- 
ence of various saccharide compounds as 
inhibitors. Shown in Table 1 are the results 
of these assays. Clearly, of the polysac- 
charides tested. Group C streptococcal 
carbohydrate was the best inhibitor of he- 



I 60 

840 

a. 

20 




2 3 4 5 6 7 8 

Log, Antiserum Dilution 

Fig. 2. Cold agglutinin activity in biphasic temper- 
ature hemolytic assay, using whole (■), Group C 
Streptococcal vaccine absorbed (•), or sheep 
erythrocyte-absorbed (A) sheep erythnx:yte antiserum. 



molysis. More than 90% inhibition was ef- 
fected in the presence of 100 ng carbohy- 
drate. At a similar concentration Type III 
pneumococcal polysaccharide was not in- 
hibitory. Of the individual sugars tested, 
^-acetylgalactosamine clearly yielded the 
most inhibition. As seen by the data in 
Table 1, approximately 200 /tA/ N-acetyl- 
galactosamine was sufficient to cause 50% 
inhibition of lysis while greater than 1200 
/jlM melibiose was required for equivalent 
inhibition activity. When lactose was tested, 
greater than 10 mAf did not cause detect- 
able inhibition of cold agglutinin antibody: 
erythrocyte interaction. Thus, lactose was 
the least effective inhibitor. Interestingly, 
galactose, although significantly less reactive 
than ^-acetylgalactosamine, was a more 
effective inhibitor than lactose (approx- 
imately 7 mAf required for 50% inhibition). 
Discussion. The demonstration that 
antibody reacts with a specific immunogen, 
implies antibody has arisen in direct re- 
sponse to the immunogen, and not indi- 
rectly as a result of nonspecific B lympho- 
cyte activation. It is possible, however, that 
under hyperimmune conditions, non-anti- 
gen-specific soluble activation factors may 
stimulate nearby lymphocytes to elicit 
a primary response. Certain cold agglutinin 
antibody responses may fall into this cate- 
gory. For example, cold agglutinin anti- 
bodies which appear during infectious ill- 
ness, or during experimental animal hyper- 
immunization can be shown to react with 



M 



C<CH4>«£ACT1VE A^rnSODY 






te-99^i 



liMiniiu^ 



S'-A^:4^fi93ditcl4f^ammie 
7>pr XIV i;aidMiy4lfal«^ 



0-22 
7.« 



No 



^ 10.0 



42 

119 



"^ flir^ mstUfhih were (s( heur0ifne€m% mokcukur weigbl: therefore, weishi %'alnes are provided. 



both h^teruMy ani^ociated and autologous 
erythrocyte-a^^ociated structures. Partially 
because of the antigenic complexity of 
these systems, it has been particularly 
difficult to determine whether (I) non- 
antigen-specific immune stimuli, (2) bac- 
teria or products in association with eryth- 
rocytes, or (3) erythrocyte cross-reactive 
bacterial determinants, have led to produc- 
tion of cold agglutinin antibodies. The pres- 
ent observation that sheep erythrocytes 
can, albeit rarely, lead to a cold agglutinin 
antibody response, suggests a cross-reac- 
tive antibody may be generated. 

It is important to note that results of the 
present work which utilized an innocuous 
agent, sheep erythrocytes, as the immuno- 
gen are similar to those obtained in other 
investigations which utilized bacterial vac- 
cines. As was observed in the present 
study, galactose-containing ligands ap- 
peared to react best with the cold agglutinin 
antibody. It is also important to note, how- 
ever, that the cold agglutinin antibody de- 
scribed in the present investigation is 
clearly different from the antibody de- 
scribed earlier in association with Group C 
streptococcal vaccine immunization (10). 
I'irst. a significantly different sensitivity to 
lactose inhibition is evident since lactose 
was not inhibitory in the present case at a 
concentration which inhibited 50% of the 
hemolytic activity of the previously de- 
scribed and characterized IgM antibody. In 
addition, these cold agglutinin antibodies 
did not discriminate between I or i determi- 
nants. Human adult and cord blood O- 
posilivc erythrocytes were equivalently 



agglutinated at 4^ Further, agglutination of 
these cells occurred at 37^ Since sheep 
erythrocytes do not carry detectable I or i 
determinants, these data suggest that reac- 
tivity with human and rabbit erythrocytes 
by these cold agglutinin antibodies does not 
involve specific discriminatory recognition 
of I or i determinants. This observation 
does not, however, exclude the possibility 
that these sugar-reactive cold agglutinin 
antibodies react with galactosyl moieties on 
the I or i oligosaccharide complex. Similar 
to the cold agglutinin-containing antistrep- 
tococcal sera reported in a previous inves- 
tigation, the single antiserum in the present 
study reacted in both the warm and cold 
with sheep erythrocytes, but only in the 
cold with rabbit erythrocytes. The Forss- 
man glycolipid glycan portion is identical 
to the immunodominant Group C strepto- 
coccal carbohydrate determinants. It ap- 
pears, therefore, that an immune response 
to Forssman antigen determinants could 
explain the source of rabbit erythrocyte 
autoreactivity. Whether other antigens were 
involved in the present response is not 
known. 

The cold agglutinin antibodies presently 
described reacted best, but not exclusively, 
with N-acetylated-a-anomers. This in- 
terpretation is based upon the data which 
clearly revealed preferential reactivity with 
iV-acetylgalactosamine versus galactose 
and melibiose versus galactose and lactose. 
Apparently, the 100%/3-anomeric structure 
present in the p 1-^4 linkage of the lactose 
molecule prevented significant reactivity 
with the antibody combining site. Further, 



COLD-REACTIVE ANTIBODY 



11 



the fact that lactose is a disaccharide per se 
is not relevant to the absence of inhibition 
observed in the presence of this compo- 
nent, since melibiose was significantly in- 
hibitory. 

That cold agglutinin antibody may appear 
during immunization with erythrocytes in 
addition to immunization with various 
bacterial vaccines, strongly suggests the re- 
sponse is specific for sheep erythrocytes 
and is independent of bacterial components 
in association with autologous erythrocyte 
membranes. These data therefore suggest 
that a clearer understanding of erythrocyte 
autoreactive responses could be attained 
through careful use of defined carbohydrate 
immunogenic stimulation. 



1. Pruzanski W, Shumak KH. Biologic activity of 
cold reacting autoantibodies. N Engl J Med 
297:538-542, 1977. 

2. Frank MM, Atkinson JP, Gadek J. Cold aggluti- 
nins and cold agglutinin disease. Annu Rev Med 
28:291-298, 1977. 

3. Schubothe H. Current problems of chronic cold 
hemagglutinin disease. Ann NY Acad Sci 
124:484-490, 1%5. 

4. Harboe M. Cold Auto-agglutinins. Vox Sang 
20:289-305, 1971. 

5. Roelcke D. Cold agglutination, antibodies and 
antigens. Clin Immunol Immunopathol 2:266- 
280, 1974. 

6. DeHeer DH, Linder EJ, Edgington, TS. Delinea- 



tion of spontaneous erythrocyte autoantibody re- 
sponses of NZB and other strains of mice. J Im- 
munol 120:825-830, 1978. 

7. Costea N, Yakulis VJ, Heller P. Experimental 
production of cold agglutinins in rabbits. Blood 
26:323-340, 1965. 

8. Costea, N, Yakulis VJ, Heller P. The mechanism 
of induction of cold agglutinins by Mycoplasma 
pneumoniae. J Immunol 106:598-604, 1971. 

9. Lind K. Production of cold agglutinins in rabbits 
induced by Mycoplasma pneumoniae. Listeria 
monocytogenes, or streptococcus MG. Acta 
Pathol Microbiol Scand Sec B 81:487-496, 1973. 

10. Colling KG, Brown JC. The appearance of IgM 
and IgG cold agglutinins in rabbits hyperim- 
munized with group C streptococcal vaccine. J 
Immunol 122:202-208, 1979. 

11. Colligan, JE, Fraser B, Kindt T. A disaccharide 
hapten from streptococcal group C carbohydrate 
that cross-reacts with the Forssman glycolipid. J 
Immunol 118:6-11, 1977. 

12. Krause RM. The search for antibodies with mo- 
lecular uniformity. Ad van Immunol 12:1-56, 
1970. 

13. Herd ZL, Spragg J. The occurrence of homogene- 
ous antibodies and selection of k light chains to 
streptococcal carbohydrates in Monash rabbits. 
Aust J Exp Biol Med Sci 50:225-243, 1972. 

14. Campbell JH, Pappenheimer AM Jr. Quantitative 
studies of antipneumococcal polysaccharide 
antibodies Types III and VIII- 1. Immunochem- 
istry 3:195-212, 1966. 



Received January 18, 1982. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 12-18 (1982) 



Effects of Culture Age on PRL and GH Responses to Bromocriptine and 
Somatostatin from Primary Cultures of Rat Anterior Pituitary Cells (41470) 

KUNIHIKO HANEW^ and EDWARD G. RENNELS^ 

Department of Anatomy, The University of Texas Health Science Center At San Antonio, 

San Antonio, Texas 78284 



Abstract. To study the effects of time in culture on prolactin (PRL) and growth hormone 
(GH) responses to exogenous stimuli, bromocriptine (10~^ M) or somatostatin (10"^ M) were 
added to primary cultures of dispersed rat anterior pituitary cells (DC). Cells which had been 
in culture for 3, 6, or 9 days were then incubated for 6 hr and the media were radioim- 
munoassayed for PRL and GH. The inhibitory effects of bromocriptine on PRL release (76% 
decrease) and somatostatin on GH release (62% decrease) from cultured cells were maximal 
on Day 3 and decreased with age of the culture. In addition, the inhibitory effect of bromo- 
criptine on GH release (32% decrease) was slight but maximal on Day 3 and decreased with 
time in culture. In contrast, somatostatin showed slight and stable inhibition of PRL release 
(26 to 29% decrease) at each of the three time periods. At the end of each incubation, the 
media were replaced with Ham*s F-10 medium lacking bromocriptine or somatostatin and 
the cultures were incubated again for 6 hr. Even after removal of these agents, the inhibitory 
effects of bromocriptine on PRL or somatostatin on GH release persisted at almost the same 
or a higher degree compared to the controls. On the other hand, the inhibitory effects of 
bromocriptine on GH or somatostatin on PRL disappeared almost completely after removal 
of the agents. The release of both PRL and GH into the media was greater in all cases, 
including the controls, during the second incubation. This may have been due in part to the 
stimulation caused by changing the media. The resuhs suggest: (1) rat pituitary mammo- 
trophs and somatotrophs may both possess receptors for bromocriptine and somatostatin (2) 
in the absence of hypothalamic control (i.e., in vitro system), these cells are most sensitive 
to both agents on Day 3, and (3) the rebound increases seen in vivo following these agents are 
likely caused by hypothalamic modulations. 



The effect of the age of primary cultures Materials and Methods. Dispersed cell 
on the course of basal hormone secretion cultures. Male Sprague-Dawley rats 
and on the reponsiveness of anterior pitu- (Charles River Corp., Wilmington, Mass.), 
itary cells to exogenous stimuli has not weighing 200-250 g, were housed in an ar- 
been well studied. In addition, the mecha- tificially illuminated (14 hr light, 10 hr dark) 
nism of rebound increases in prolactin animal room for periods of 1 to 2 weeks 
(PRL) and growth hormone (GH) secretion before sacrifice. Food and water were 
in rats and humans after cessation of treat- available ad libitum. Twenty-three animals 
ment with inhibitory agents is not fully were killed at 09:00 hr by decapitation. The 
understood (1-7). To investigate these pituitary glands were removed aseptically 
important questions, we used primary cul- under a laminar flow hood, the posterior 
tures of rat anterior pituitary cells and ex- lobes were discarded, and the remaining 
amined the basal secretion and the respon- tissue was placed into a sterile petri dish 
siveness of the mammotrophs and somato- (5-cm diameter) and minced to the size of 1 
trophs to bromocriptine and somatostatin mm^ in Hank's balanced salt solution with- 
on Days 3, 6, and 9 in culture. out Ca^^ or Mg^"^. The minced pituitary tis- 
sue was washed gently with Hank's solu- 

• Present address: The Second Department of In- tion five times to remove blood cells. Then 

ternal Medicine, Tohoku University School of Med- 5 ml of 0.1% trypsm (Grand Island Biologl- 

icine, 1-1, Seiryo-cho, Sendai 980, Japan. cal Co., GIBCO, Grand Island, N.Y.), sol- 

^ To whom reprint requests should be addressed. ubilized in Hank's solution containing 0.1% 

12 
00-^7-9727/82/0900 1 2-07$0 1 .00/0 

fht ^^ 1982 by the Society for Experimental Biology and Medicine. 
/» reaervcd. 



PRL AND GH SECRETION In Vitro 



13 



bovine serum albumin (BSA; Sigma Chem- 
ical Co., St. Lx>uis, Mo.) was added to the 
petri dish, and the tissue was then trans- 
ferred to a 35-ml Erlenmeyer flask. The 
solution was agitated gently by a stirrer 
for 10 min, and the supernatant, which 
contained dispersed pituitary cells (DC), 
was collected by filtration through lens 
paper (Matherson Scientific , New York) in 
an ice-cooled 50-ml Erlenmeyer flask. 
Then, 5 ml of Ham's F-10 medium sup- 
plemented with 10% fetal calf serum and 
2.5% horse serum was added to the flask to 
block the action of the trypsin. The tryp- 
sinization and cell collection procedures 
were repeated five times. Finally, the cell 
suspensions were centrifuged for 5 min at 
1000 rpm, and the cells were resuspended in 
Ham's F-10 medium. These cells were cul- 
tured in T-25 flasks (Coming, No. 25100) at 
37** for 3, 6, or 9 days in an atmosphere of 
5% CO2 and 95% air. The number of cells 
per flask was 16.5 x W, and the cell viabil- 
ity, as determined by the trypan blue dye 
exclusion method, was over 95% before 
and after 9 days of cell culture. 

Effects of somatostatin or bromocriptine 
on PRL and GH release. After 3, 6, or 9 
days of cell culture, the medium was dis- 
carded and 10"^ M of somatostatin (GH re- 
lease inhibiting hormone; GH-IH, Sigma) 
or 10"^ M of bromocriptine (2-Br-a-ergo- 
criptine; CB-154, Sandoz) dissolved in 
Ham's F-10 medium lacking serum (pH 7.7) 
was added to the DC cultures. Cells cul- 
tured in serum-free Ham's F-10 medium 
were used as controls. Five flasks were 
used for the control cultures or for the 10"^ 
M concentration of each agent, respec- 
tively. After 6 hr in the presence of these 
agents, the media were collected. These 
samples were kept frozen at -27° until as- 
sayed. 

PRL and GH secretory responses after 
the removal of the agents. After the collec- 
tion of the media mentioned above, 5 ml of 
Ham's F-10 medium lacking serum and 
agents were added to every flask, and the 
cells were again incubated for 6 hr to see if 
rebound increases in the hormones released 
would occur in vitro. The media were then 
collected and kept frozen until assay. In 



both experiments, the effects produced by 
the agents were compared to the results 
obtained in control cultures. 

RIA of rat PRL and GH, Rat PRL was 
measured in duplicate by radioimmunoas- 
say (RIA) methods previously described 
(8). Rat GH was also measured by double 
antibody RIA procedures using materials 
supplied by the NIAMDD. All samples 
were assayed in duplicate according to the 
procedures recommended by NIAMDD. 
Results were expressed in terms of the NIH 
standards, PRL-RP-1 and GH-RP-1. Intra- 
and interassay coefficients of variation 
were 2.1 and 4.8% for PRL, and 5.9 and 
7.5% for GH, respectively. 

Statistical analysis. The data were sub- 
jected to analysis of variance followed by 
the Student-Newman-Keul's test for 
comparing the differences between group 
means. 

Results. Effects of somatostatin or 
bromocriptine on PRL release. The basal 
secretion rate of PRL increased progres- 
sively with increasing time in culture (Fig. 
lA). At each of the three time periods not 
only bromocriptine but also somatostatin 
significantly suppressed the rate of PRL 
release compared to the control values {P < 
0.01; Fig. lA). When these data were ex- 
pressed as a percentage of the control value 
(Fig. IB) the inhibitory effect of somato- 
statin on PRL release was constant and 
varied from 26 to 29% at the three time 
periods. In contrast, the inhibitory effect of 
bromocriptine on PRL release was maximal 
on Day 3 (76% decrease from basal), and 
became attenuated at 6 days and at 9 days 
(52 and 32% decreases in PRL release from 
basal on Days 6 and 9, respectively). The 
inhibition by bromocriptine was far greater 
than that produced by somatostatin on Day 
3 (P < 0.01) and on Day 6 (P < 0.01), but 
not on Day 9 (Fig. lA). 

PRL secretory responses after removal 
of the agents. After removal of the media 
containing the inhibitory agents the DC 
were reincubated for 6 hr in Ham's F-10 
medium alone. Again the control flasks 
showed increases in PRL release with in- 
creasing time in culture and each value ex- 
ceeded that of the former incubation (Fig. 



14 



PRL AND GH SECRETION In Vitro 





6 9 "36 

Days Days 

Fig. 1 . (A) Effects of somatostatin (GH-IH) and bromocriptine (CB- 154) on the release of PRL into 
Ham's F-10 culture medium during a 6-hr incubation. Each bar represents the mean ± SEM . (B) The 
mean values were expressed as percentage of control (100%). 



2A). Even after the removal of the bromo- 
criptine from the cultures the PRL inhib- 
itory effect of this agent continued at almost 
the same degree as was seen with bromo- 
criptine in the medium (71, 56, and 52% de- 
crease from basal on Days 3, 6 and 9, re- 
spectively; P < 0.01 compared to control). 
In contrast, the removal of somatostatin 



from the media largely freed the cells from 
its inhibitory effect; only in the 9-day cul- 
tures was there still a significant inhibition 
of PRL release as compared to the control 
group (P < 0.01; Figs. 2A, B). 

Effects of somatostatin or bromocriptine 
on GH release. Basal GH release rate in- 
creased with time in culture and reached a 





Fig. 2. (A) PRL secretory responses during 6-hr incubation with agent-free Ham's F-10 culture 
medium after the removal of somatostatin (GH-IH) and bromocriptine (CB-154). (B) See legend of 



PRL AND GH SECRETION In Vitro 



15 





Fig. 3. (A) Effects of somatostatin (GH-IH) and bromocriptine (CB-154) on the release of GH into 
Ham^s F-10 culture medium during a 6-hr incubation. (B) See legend of Fig. IB. 



plateau at 6 days (Fig. 3A). At each time, 
somatostatin and bromocriptine signifi- 
cantly inhibited the GH release compared 
to controls (P < 0.05-0.01). As shown in 
Fig. 3B, the inhibitory effects of somato- 
statin or bromocriptine on GH release were 
maximal on Day 3 (62 and 32% decrease, 
respectively) and were diminished on Day 6 
(23 and 20% decrease) and on Day 9 (22 and 
17% decrease, respectively). On Day 3, the 
inhibitory action of somatostatin was far 
greater than that of bromocriptine (P < 
0.01). However, there were no significant 
differences between the inhibitory effects 
of somatostatin and bromocriptine on Days 
6 or 9. 

GH secretory responses after the re- 
moval of the agents. After removal of the 
inhibitory agents, the GH released in the 
control group was maximal on Day 3, and 
lessened with time although all of the basal 
control values were far greater than those 
seen in the first incubation (Fig. 4A). As 
with the PRL responses to bromocriptine, 
after the removal of somatostatin, the inhib- 
itory effect on GH release persisted at a 
slightly higher level than that seen during 
the first incubation (67, 50, and 49% de- 
crease on Days 3, 6, 9, respectively, P < 



0.01 compared to control; Figs. 4A, B). On 
the other hand, the cultures that had been 
treated with bromocriptine were freed from 
inhibition and showed no values which 
differed significantly from the controls 
(Figs. 4A, B). 

Discussion. This study showed that 
bromocriptine or somatostatin can mark- 
edly suppress the rate of PRL and GH re- 
lease from primary cultures of dispersed rat 
anterior pituitary cells. With increasing 
time in culture the PRL- or GH-inhibiting 
activity of bromocriptine or somatostatin, 
respectively, became somewhat decreased. 
Moreover, we observed no rebound in- 
creases in the release of either hormone 
after the removal of these inhibitory agents. 

It has been reported that dopamine (DA) 
or DA agonists can suppress PRL and GH 
secretion in adult rats (9-15), although DA 
may have a stimulatory role on GH secre- 
tion in infant rats (16). In this study, 
bromocriptine had an inhibitory effect not 
only on PRL but also on GH release in 
vitro. Thus bromocriptine acts on pituitary 
mammotrophs and somatotrophs directly. 
Therefore, it seems possible that rat pitu- 
itary mammotrophs and somatotrophs both 
have receptors for bromocriptine. It has 



16 



PRL AND GH SECRETION in Vitro 





6 9 '36 

Days Days 

Fig. 4. (A) GH secretory responses during 6-hr incubation with agent-free Ham's F-10 culture 
medium after the removal of somatostatin (GH-IH) and bronnocriptine (CB-154). (B) See legend of 
Fig. IB. 



been demonstrated that receptors for DA 
and apomorphine are both present in cells 
of the pituitary gland (17-19). 

In rats, somatostatin suppresses GH 
secretion both in vivo and in vitro, whUe 
it inhibits PRL secretion only in vitro 
(20-22). Our results confirmed the in vitro 
effects of somatostatin on GH and PRL re- 
lease. The reason somatostatin inhibits 
PRL release only in vitro is not known. One 
possibility may be that somatostatin can re- 
veal this action only in the absence of 
hypothalamic regulation. In normal human 
subjects, somatostatin does not inhibit 
basal or TRH-induced PRL secretion (23, 
24). However, it inhibits the basal levels of 
PRL in some acromegalic patients (25) and 
arginine-induced PRL release in normal 
men (26). In acromegalic patients, the 
properties of mammotrophs and the hypo- 
thalamic function of the tuberoinfiindibular 
dopaminergic neurons may be different 
than in normal subjects (25, 27, 28). 

Studies on the time course of the respon- 
siveness of primary cultures of dispersed 
anterior pituitary cells to bromocriptine or 
somatostatin have not been previously re- 
ported. Bromocriptine had maximal inhib- 
iting activites on PRL or GH release on Day 



3, and this activity became lessened with 
time. Somatostatin caused a maximal inhi- 
bition of GH release on Day 3, and this in- 
hibition decreased with time. In contrast, 
somatostatin showed a slight but stable in- 
hibition of PRL release at each of the three 
times tested. Why bromocriptine and 
somatostatin had maximal inhibitory ac- 
tivities on PRL or GH release on Day 3, 
respectively, and then showed a lesser ac- 
tivity with increasing time in culture is not 
clear. These findings are at variance with 
the reports that surgical or pharmacological 
disruption of the hypothalamus causes 
supersensitivities of mammotrophs or so- 
matotrophs to exogenous stimuli (29- 
31). However, in these experiments the 
pituitary glands were already manipu- 
lated in vivo, and the in vitro studies 
employed hemipituitary glands obtained 
immediately after decapitation. Therefore, 
the experimental conditions were quite 
different from ours and it is difficult to 
compare the results. Without the presence 
of receptor binding substances, it is possi- 
ble that the number of receptors to regulat- 
ing hormones and responsiveness of the 
cells will decrease, as has been suggested 
for patients with Addison's disease where 



PRL AND GH SECRETION In Vitro 



17 



the high ACTH levels are not suppressed to 
normal ranges by glucocorticoids (32, 33). 

Somatostatin and DA are important al- 
though not the only physiological regulators 
of GH and PRL secretion, respectively (9, 
13, 20, 34). Bromocriptine is a potent DA 
agonist, and has been widely used as a sub- 
stance to inhibit PRL release (13, 35). Al- 
though, the responsiveness of GH to 
somatostatin or PRL to bromocriptine 
changed markedly with time in our experi- 
ments those of GH to bromocriptine or 
PRL to somatostatin showed only slight 
changes. These findings may also suggest 
that, in the absence of hypothalamic con- 
trol, somatotrophs and mammotrophs may 
lose their responsiveness to their major in- 
hibiting agents, while the cells may retain 
their responsiveness to other less specific 
agents. 

In vivo studies have shown that after the 
removal of inhibiting agents, PRL or GH 
are apt to show rebound increases exceed- 
ing the initial basal levels (1-7). We did not 
observe any rebound phenomenon in PRL 
or GH release after the removal of the in- 
hibitory agents, although the level of hor- 
mones released from all the cultures, in- 
cluding the controls, was higher after the 
removal of somatostatin or bromocriptine. 
We believe these increases as well as the 
higher basal levels of hormone release 
must have been due to at least in part to the 
stimulatory effect of changing the culture 
media (36, 37). 

There are several possible explanations 
for the mechanism of the post-inhibitory 
rebound seen in vivo. The rebound could be 
due to (1) increased hypothalamic releasing 
factors; (2) decreased hypothalamic inhib- 
iting factors; or (3) an overshoot from the 
stored pituitary pool (3, 5). Our results 
mainly support the first or second explana- 
tions because without the hypothalamic 
control in vitro there were no rebound 
phenomena. These fmdings are similar to 
those of Martin et al. in which hypothal- 
amic VMN (ventromedial nucleus) lesions 
in rats prevented the somatostatin-induced 
postinhibitory rebound (3). 

In conclusion, rat pituitary mammo- 
trophs and somatotrophs have been shown 
to respond in vitro to both bromocriptine 
and somatostatin. With the removal of 



hypothalamic control (i.e., in vitro system), 
these cells were most sensitive to these in- 
hibitory agents on Day 3 in culture. Also, it 
is proposed that the post-treatment rebound 
increases exceeding basal levels in PRL and 
GH secretion seen in vivo must be mainly 
caused by hypothalamic modulations. 

The rat PRL and GH kits used in the radioim- 
munoassays were kindly provided by NIAMDD, 
through the courtesy of Dr. A. F. Parlow. This re- 
search was supported by USPHS Grant AM 12583. 

1. Blake CA. Effects of intravenous infusion of cate- 
cholamines on rat plasma luteinizing hormone and 
prolactin concentrations. Endocrinology 98:99- 
104, 1976. 

2. Levin JI, Voogt JL. Effects of dopamine, norepi- 
nephrine, and serotonin on plasma prolactin levels 
inovariectomized, pituitary-grafted rats. Proc Soc 
Exp Biol Med 157:576-579, 1978. 

3. Martin WH, Rogol AD, Kaiser DL, Thomer MO. 
Dopaminergic mechanisms and luteinizing hor- 
mone (LH) secretion. IL Differential effects of 
dopamine and bromocriptine on LH release in 
normal women. J Clin Endocrinol Metab 
52:650-656, 1981. 

4. Martin JB. Brain regulation of growth hormone 
secretion. In: Martini L, Ganong WF, eds. Fron- 
tiers in Neuroendocrinology. New York, Raven 
Press, pl29, 1976. 

5. Stachura ME. Influence of synthetic somatostatin 
upon growth hormone release from perifiised rat 
pituitaries. Endocrinology 99:678-683, 1976. 

6. Malven PV. Immediate release of prolactin and 
biphasic effects on growth hormone release fol- 
lowing electrical stimulation of the median em- 
inence. Endocrinology 97:808-815, 1975. 

7. Leblanc H, Rigg LA, Yen SSC. The response of 
pancreatic and pituitary hormones to pulses and 
constant infusion of somatostatin. J Clin Endo- 
crinol Metab 41:1105-1109. 1975. 

8. Hanew K, Shiino M, Rennels EG. Effects of in- 
doles, AVT, oxytocin, and AVP on prolactin se- 
cretion in rat pituitary clonal (2B8) cells. Proc Soc 
Exp Biol Med 164:257-261, 1980. 

9. Gibbs DM, Neill JD. Dopamine levels in 
hypophysial stalk blood in the rat are sufficient to 
inhibit prolactin secretion in vivo. Endocrinology 
102:1895-1900, 1978. 

10. Gudelsky GA, Porter JC. Release of newly syn- 
thesized dopamine into the hypophysial portal 
vasculature of the rat. Endocrinology 104:583- 
587, 1979. 

11. Yeo T, Thomer MO, Jones A, Lowry PJ, Besser 
GM. The effect of dopamine, bromocriptine, ler- 
gotrile and metoclopramide on prolactin release 
from continuously perfused columns of isolated 



18 



PRL AND GH SECRETION In Vitro 



rat pituitary cells. Clin Endocrinol 10:123-130, 
1979. 

12. Smalstig EB, Sawyer BD, Clemens JA. Inhibition 
of rat prolactin release by apomorphine in vivo 
and in vitro. Endocrinology 95:123-129. 1974. 

13. MacLeod RM. Regulation of prolactin secretion. 
In: Martini L, Ganong WF, eds. Frontiers in 
Neuroendocrinology. New York, Raven Press, 
pl69, 1976. 

14. Collu R, Frashini F, Visconti P, Martini L. Ad- 
renergic and serotonergic control of growth hor- 
mone secretion in adult male rats. Endocrinology 
90:1231-1237, 1972. 

15. Kokka N, Garcia JF, Elliott HW. Effects of acute 
and chronic administration of narcotic analgesics 
on growth hormone and corticotrophin (ACTH) 
secretion in rats. Prog Brain Res 39:347-360, 
1973. 

16. Mflller EE, Nistico G, Scapagnini U. Neuro- 
transmitters and Anterior Pituitary Function. 
New York, Academic Press, p246, 1977. 

17. MacLeod RM, Kimura H, Login I. Inhibition of 
prolactin secretion by dopamine and piribedil 
(ET-495). In: PecUe A, MiiUer EE, eds. Growth 
Hormone and Related Peptides, Amsterdam, Ex- 
cerpta Medica p443, 1976. 

18. Caron MG, Beaulieu M, Raymond V, Gagn^ B, 
Drouin J, Lefkowitz RJ, Labrie F. Dopaminergic 
receptors in the anterior pituitary gland. J Biol 
Chem 253:2244-2253, 1978. 

19. Cronin MJ, Robert JM, Weiner RI. Dopamine and 
dihydroergocriptine binding to the anterior pitu- 
itary and other brain areas of the rat and sheep. 
Endocrinology 103:302-309, 1978. 

20. Brazeau P. Vale W, Burgus R, Ling N, Butcher 
M, Rivier J, Guillemin R. Hypothalamic polypep- 
tide that inhibits the secretion of immunoreactive 
pituitary growth hormone. Science 179:77-79, 
1973. 

21. Vale W, Rivier C, Brazeau P, Guillemin R. Effects 
of somatostatin on the secretion of thyrotropin 
and prolactin. Endocrinology 95:968-977, 1974. 

22. Drouin J, L^an AD, Rainville D, Lachancc R, 
Labrie F. Characteristics of the interaction be- 
tween thyrotropin-releasing hormone and 
somatostatin for thyrotropin and prolactin re- 
lease. Endocrinology 98:514-521, 1976. 

23. Siler TM, Vandenberg G, Yen SSC, Brazeau P, 
Vale W, Guillemin R. Inhibition of growth hor- 
mone release in humans by somatostatin. J Clin 
Endocrinol Metab 37:632-634, 1973. 

24. Hall R, Besser GM, Schally AV, Coy DH, Evened 

C, Goldie DJ, Kastin AJ, McNeilly AS, Mortimer 
CH, Phenekos C, Tunbridge WMG, Weightman 

D. Action of growth-hormone-release inhibiting 
hormone in healthy men and in acromegaly. Lan- 
cet 2:581-584, 1973 

25. Yen SSC, Siler TM, De Vane GW. Effect of 
somatostatin in patients with acromegaly. Sup- 



pression of growth hormone, prolactin, insulin 
and glucose levels. N Engl J Med 290:935- 
938, 1974. 

26. Rodriguez-Amao MD, Gomez-Pan A, Rainbow 
SJ, Woodhead S, Comaru>Schally AM, Schally 
AV, Meyers CA, Coy DH, Hall R. Effects of pro- 
somatostatin on growth hormone and prolactin re- 
sponse to arginine in men. Comparison with 
somatostatin. Lancet 1:353-356, 1981. 

27. Cryer PE, Daughaday WH, Growth hormone. In: 
Martini L, Besser GM, eds. Clinical Neuroendo- 
crinology. New York, Academic Press, p243, 
1977. 

28. Collu R. Role of central cholinergic and aminergic 
neurotransmitters in the control of anterior pitu- 
itary hormone secretion. In: Martini L, GM 
Besser, eds. Clinical Neuroendocrinology. New 
York, Academic Press, p43, 1977. 

29. Cheung CY, Weiner RI. In vitro supersensitivity 
of the anterior pituitary to dopamine inhibition of 
prolactin secretion. Endocrinology 102:1614- 
1620, 1978. 

30. Anunziato L, Moore KE. Increased ability of 
apomorphine to reduce serum concentrations of 
prolactin in rats treated chronically with a- 
methyltyrosine. Life Sci 21:1845-1850, 1977. 

31. Udeshini G, Cocchi D, Panerai AE, Gil- Ad I, 
Rossi GL, Chiodini PG, Liuzzi A, Miiller EE. 
Stimulation of growth hormone release by 
thyrotropin-releasing hormone in the hypophy- 
sectomized rat bearing an ectopic pituitary. En- 
docrinology 98:807-814, 1976. 

32. Holdaway IM. Suppression of plasma ACTH 
levels with corticosteroids in Addison's disease. 
Clin Endocrinol 2:37-42, 1973. 

33. Krieger DT, Gewirtz GD. The nature of the circa- 
dian periodicity and suppressability of im- 
munoreactive ACTH levels in Addison's disease. 
J Clin Endocrinol Metab 39:46-52, 1974. 

34. Chihara K, Arimura A, K-Garfias C, Schally AV. 
Enhancement of immunoreactive somatostatin 
release into hypophysial portal blood by electrical 
stimulation of the preoptic area in the rat. Endo- 
crinology 105:1416-1418, 1978. 

35. Corrodi H, Fuxe K, Hokfelt T, Lidbrink P, Un- 
gerstedt U. Effect of ergot drugs on central cate- 
cholamine neurons: Evidence for a stimulation of 
central dopaminergic neurons. J Pharm Phar- 
macol 25:409-411, 1973. 

36. Groshong JC, Milo GE, Malarkey WB. The effect 
of aging on prolactin secretion in rat pituitary 
monolayer cuhure. Life Sci 20:1821-1828, 1977. 

37. Kim KC, Burkman AM. Prolactin release from rat 
anterior pituitary cell culture following medium 
replacement. Proc Soc Exp Biol Med 167:551- 
553, 1981. 

Received December 21, 1981. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 19-23 (1982) 

The Transport of Urate in the Small Intestine of the Rat (41471) 

C. E. DUKES, D. A. STEPLOCK, A. M. KAHN, and E. J. WEINMAN* 
Division of Nephrology, University of Texas Medical School, P.O. Box 20708, Houston. Texas 77025 



Abstract. The transport of urate in the small intestine of the rat was examined, in vivo, to 
determine if specific (mediated) transport systems arc present. The secretory flux of urate 
was determined following the infusion of urate systemically, with or without oxanate, while 
the small intestine was perfiised with an initially urate-free solution. The absorptive flux of 
urate was determined by perflising the gut with solutions of known concentrations of urate. 
In some studies probenecid was infused systemically or added to the luminal perfusion 
solution. Over a wide range of concentrations of urate in the plasma or in the lumen, there 
was no evidence for saturation of either the secretory or absorptive fluxes of urate. Pro- 
benecid had no effect on either of the flux rates. These flndings suggest that in the small 
intestine of the rat, the movement of urate out of or into the lumen occurs by passive 
diffusion and that, under the conditions of study, no evidence for facilitated transport can be 
demonstrated. 



In previous studies from this laboratory, 
we have examined, in detail, the mecha- 
nisms subserving the excretion of urate by 
the kidney (1, 2). In the rat kidney, the 
transport of urate in both the reabsorptive 
and secretory directions are mediated 
transport processes (1, 2). Despite general 
agreement that the kidney is the major 
route of excretion of urate, there is ample 
data to indicate that the gastrointestinal 
tract also plays a role in the overall metab- 
olism of urate (3). The nature of the trans- 
port of urate in the gastrointestinal tract, 
however, has not been clearly defmed. In 
the present studies we have examined the 
rates of transfer of urate out of and into the 
small intestine of the rat in order to deter- 
mine if specific transport systems for this 
organic anion are present. 

Methods. Male Sprague-Dawley rats 
with free access to food and water prior to 
study were anesthetized with pentabarbital 
(50 mg/kg body wt injected intraperitone- 
ally). A tracheotomy was performed, the 
femoral artery and vein were cannulated 
and the urinary bladder was catheterized. 
The abdominal cavity was opened via a 



' To whom correspondence should be addressed. 



midline incision and segments of the small 
intestine were cannulated with polyethyl- 
ene catheters. Proximal segments of the 
small intestine were considered to be from 
the ligament of Trietz and extending distally 
9 to IS cm. Distal segments of the small 
intestine were considered to be 5 cm from 
the ileocecal valve and extending proxi- 
mally 9 to 15 cm. Animals were placed on a 
heated board and body temperature was 
maintained at 'iV, During the course of 
study each animal received an intravenous 
infusion of isotonic saline at a rate of 6 
ml/hr. The small intestines were perfused at 
a rate of 12 ml/hr with a solution containing 
sodium 130 meq/liter, bicarbonate 30 meq/ 
liter, potassium 5 meq/liter, and chloride 
105 meq/liter. [^HJPolyethylene glycol 
(PEG) (New England Nuclear Corp., Bos- 
ton, Mass.) was added as a volume marker. 
The collection periods were 10 to 20 min in 
duration. 

In 30 animals the secretory flux of urate 
was determined by the systemic infusion of 
urate with or without oxanate, an inhibitor 
of uricase, while perfusing the intestine 
with an initially urate free solution. In order 
to obtain a range of plasma concentrations 
of urate, sodium urate in concentrations of 
10, 25, or 50 mg% and oxanate in concen- 
trations of 0.5 or 1.0 g% were added to the 



19 
0037-9727/82/090019-05$01 .00/0 

AW n|hx« TeMTv«A. 



20 



URATE TRANSPORT IN RAT SMALL INTESTINE 



intravenous infusion solution. Where ex- 
amined, probenecid was also added to in- 
travenous infusion solution in concentra- 
tions calculated to deliver a dose equal to 
100 mg/kg body wt/hr. In other studies, 
probenecid (28.5 mg%) was added to the 
intestinal perfusion solution. 

In the studies examining the absorptive 
flux of urate, no urate or oxanate was in- 
fused systemically (20 animals). Sodium 
urate was added to the intestinal perfusion 
solution in concentrations of 5 to 20 mg%. 
Where examined, probenecid was infused 
systemically or added to the luminal perfu- 
sion solution in the amounts previously in- 
dicated. 

In both the secretory and absorptive flux 
studies, two intestinal perfusions were ob- 
tained at any given intravenous infusion 
rate or any given initial concentration of 
urate in the perfusion solution. The urate 
concentrations were then changed, 20 ad- 
ditional min allowed to elapse to allow for a 
new steady state and additional collections 
obtained. Usually three such periods at 
different concentrations of urate were ob- 
tained per animal. At the end of the experi- 
ments, the distance between the perfusion 
and collection intestinal catheters was mea- 
sured. The average length of perfused seg- 
ments of the small intestine averaged 11.6 
± 0.05 cm for the proximal segments and 
11.4 ± 0.61 cm for the distal segments. 

The radioactivity of perfusion and col- 
lection solutions was determined in Bio- 
fluor (New England Nuclear Corp.) in a liq- 
uid scintillation counter. The concentration 
of urate was determined by high perfor- 
mance liquid chromatography with electro- 
chemical detection as previously described 
(4). Neither oxanate or probenecid interfered 
with the determination of the concentra- 
tions of urate. 

The in vivo perfusion rate was calculated 
from the formula: perfusion rate (ml/hr) = 
collected volume (ml/hr) x CF/PFpEc; where 
CF and PF are the disintegrations per min- 
ute of PEG in the collected fluid (CF) and 
perfusion solution (PF), respectively. The 
secretory flux of urate was calculated from 

is (g/min/cm) = CF urate x collected 
volume X length" \ 



where CF urate is the concentration of 
urate in the collected perfusion solution, 
and length is the distance between perfu- 
sion and collection sites in centimeters. The 
absorptive flux was calculated from 

Ja (g/min/cm) = {(1-(CF/PF urate)/ 
(CF/PFpeg)} 
X perfusion rate 
X PF urate x length"* 

where PF urate is the concentration of urate 
in the inital perfusion solution. The results 
of the duplicate collections were averaged 
and the results presented as the mean of 
means ± SEM. 

Results. The CF/PFpeg, an index of 
water absorption averaged 1 .04 ± 0.004 and 
1.06 ± 0.0()6 for proximal and distal seg- 
ments, respectively. The perfusion rates 
were also similar in both intestinal seg- 
ments and averaged 12.4 ± 0.09 ml/hr for 
proximal segments and 13.4 ± 0.21 ml/hr 
for distal segments. The rates or water ab- 
sorption were not effected by the presence 
of urate or probenecid in the lumen or by the 
systemic irfusion of urate or probenecid. 

For the sake of data presentation, the 
fluxes are averaged for urate concentra- 
tions over a range of 1 mg% for plasma 
urate concentrations from less than 1 to 5 
mg% and in a 3 mg% intervals for higher 
plasma concentrations of urate. There were 
no significant differences in the rates of 
urate secretion in proximal and distal intes- 
tinal segments at any given plasma concen- 
tration of urate. In addition, neither the 
systemic infusion of probenecid nor the in- 
clusion of probenecid in the luminal perfu- 
sion solution had any effect on the secretion 
of urate. The results of all secretory studies 
were combined and are plotted as a func- 
tion of the plasma concentration of urate 
(Fig. 1). As can be seen, over the wide 
range of plasma concentrations of urate ob- 
tained, the secretion of urate appeared to be 
a direct function of the plasma concentra- 
tion and no evidence for saturation of the 
secretory flux is evident. 

The absorptive flux of urate was also 
similar in both proximal and distal segments 
and was not influenced by the inclusion of 
probenecid in the perfusion solution. Figure 



UKATE T1t%NSPOKT IN lt\T SdC\LL INTESTINE 



II 



< tab 
5 • 

-A. 



i*^ 



^t 



2^ 5« 



''a 2 4 C 8 10 12 14 IS U 20 

UMTE CQMC»T1UTK)M im^} 

Fig. I. Tke reiitiottship between the secretory flux 
of orale icloscd circles) or the absorptive flux of urate 
(open circles) aod either the plasma concentration of 
orale in the case of the secretory studies or the arith- 
metic mean intralumiiiai concentration of urate In the 
absorptive studies. Values represent mean of means ± 
SEM. Each point b the average of 4 to 15 experimental 
periods. 

1 suminarizes the restilts of these studies 
and illustrates the absorptive flux as a 
function of the arithmetic mean intralumi- 
nal concentration of urate. In these studies, 
the plasma concentration of urate was less 
than 1 mg%. The values illustrated were 
grouped in a manner identical to that of the 
secretory studies. As was the case in the 
secretory studies, the absorptive flux is a 
direct function of the mean intraluminal 
concentration of urate and no evidence for 
saturation of the absorptive flux is evident. 
At any given concentration of urate, the ab- 
sorptive flux approximates the secretory 
flux. 

Discussion. It is known that the inges- 
tion of a high purine diet increases the uri- 
nary excretion of urate and, in some indi- 
viduals, exacerbates hypenirecemia. This 
response presumably reflects absorption of 
urate precursors which are metabolized to 
urate in the liver. It has also been demon- 
strated that urate itself can be reabsorbed 
from the gastrointestinal tract (3). The 
question of gut absorption of urate might be 
rendered moot when it is appreciated that 
diet of neither man nor rat contains any 
preformed urate. On the other hand, the gut 
is thought to dispose of 25 to 33% of the 
daily production of urate (3). Reabsorption 
of this component of urate in the lumen 
could be of potential importance. The 
mechanism whereby urate enters the intes- 



tine lisis not heten cburifk^. Cc>nceix«N>x 
umie c<>ukl enter ;ju$^ v^>mpcM(ieQl^ o4r ysfts- 
iroinleslinat secretions. piji:$^\i^ly ditt\i$)e 
from the Mood into the ](i;i:i$^ln>inte:i$^lin«l 
tract, or be secreted by specific mealed 
tninspofft systems analogous to Iho^ pres- 
ent in the kidney (K 2K 

Kobiss;t ri al. have recently presented 
evidence that hypoxMilhine ;uid XMilhine 
are actively secreted by the jt^unum of the 
guinea pig (5>. The transport of urate itself, 
however, was not investigated in these 
studies. Berlin and Hawkins have also re- 
ported active secretion of hypoxanthine 
and xanthine in the small intestine of the 
golden hamster (16). Although less defini- 
tive than for hypoxanthine and xanthine, 
these studies also suggested that the carrier 
had specificity for urate as well. Scharrer fi 
al. have demonstrated an active absorptive 
flux for hypoxanthine in the jejunum of the 
lamb (7). The transport of urate was not 
specifically examined in these studies but 
urate had no effect on the transport of 
hypoxanthine. Active absorption of some 
purines and pyrimidines has also been 
demonstrated in the chiton but no evidence 
for the active absorption of urate was found 
in this species (8). 

Shanker et al. demonstrated that the rat 
intestine was capable of active absorption 
of uracil and that certain purine and 
purine-like compounds could inhibit its 
transport (9). Although it was not possible 
to demonstrate unequivocally that urate it- 
self was transported, urate did inhibit, al- 
beit weakly, the transport of uracil sug- 
gesting a common transport mechanism. 
Wilson and Wilson, Kahn et uL, and Oh tt 
aL, however, have all presented evidence 
that urate transport across the small inten- 
tine of the rat was by passive diffusion only 
and no active or mediated transport Myn- 
tems could be discerned (10-12). The va- 
lidity of the conclusions of these Htudies has 
been questioned in some of these experi- 
ments since the studies were performed /// 
vitro and no data were presented to indicate 
that the tissues examined were capable of 
active transport of other subHtrates (5). 
Moreover, these studies employed |2- 
'^CJurate which can conceivably undergo 



22 



URATE TRANSPORT IN RAT SMALL INTESTINE 



metabolic transformation to radioactive al- 
lantoin. 

Since some uncertainty as to whether or 
not the intestine of the rat is capable of 
transport of urate, and since the intestine 
may be important in the overall disposition 
of urate, we elected to reinvestigate this 
question using the techniques of in vivo in- 
testinal perfusion. In the rat, urate is con- 
verted to allantoin and, as a consequence of 
this conversion, the infusion of urate sys- 
temically does not lead to predictable ele- 
vations in the serum concentration of urate. 
Moreover, renal function was intact in our 
animals and no doubt the urinary excretion 
of urate was increased. However, by infusing 
varying amounts of urate with or without 
oxanate, a hepatic inhibitor of uricase, we 
were able to study the recovery of urate in 
perfused segments of the gut in response to 
a wide range of plasma concentrations. The 
secretory flux of urate into the small intes- 
tine, at any given plasma concentration of 
urate, was similar in proximal and distal 
segments of the small intestine. As summa- 
rized in Fig. 1 , no clear evidence of satura- 
tion of the secretory process was evident 
although some flattening of the relationship 
is noted at the highest plasma concentra- 
tions of urate obtained. It might be argued 
that a low capacity transport system does 
exist and was nearly saturated even at 
the lowest concentrations examined. It is 
also possible that a carrier system is pres- 
ent, one which has a K„, above that of the 
plasma concentrations obtained in the pres- 
ent study. Against these suggestions is the 
fact that probenecid in concentrations or 
amounts known to inhibit the renal trans- 
port of urate, had no influence on transport 
in the small intestine. To the degree that 
probenecid is a reasonably specific inhib- 
itor of organic anion transport in a variety 
of tissues and based upon the relationship 
between the secretory flux and the plasma 
concentration, it seems reasonable to con- 
clude that urate enters the lumen of the in- 
testine by a passive mechanism and that a 
specific mediated transport system cannot 
unequivocally be demonstrated. We have 
also investigated the absorptive flux of 
urate over a range of concentrations of 



urate in the lumen. The imposition of a 
lumen to blood gradient for urate results in 
absorption of luminal urate. There was a di- 
rect relationship between the average lumi- 
nal concentration and the absorptive flux 
over a wide range of urate concentrations. 
There was no clear evidence however for 
saturation of the absorptive process. 
Moreover, as in the secretory studies, the 
absorptive flux of urate was not influenced 
by the infusion of probenecid systemically 
or the inclusion of probenecid in the luminal 
solution. Using analogous considerations to 
those already discussed, it seems likely that 
the absorption of urate is also occurring by 
passive permeation only. The conclusions 
about both the nature of the secretory and 
the absorptive flux are entirely consistent 
with some of the evidence already cited 
from studies performed in the rat (10-12). 
It is possible that species differences ac- 
count for some of the disparities in the liter- 
ature and it is recognized that interspecies 
variability in the renal handling of organic 
anions exists. It is likely that such may also 
exist in the intestinal transport of organic 
anions. It seems reasonable to conclude 
from the present studies plus the studies 
performed by others that in the small intes- 
tine of the rat, urate transport is most likely 
due to passive diffusion. The present 
studies, however, do not rule out the possi- 
bility of mediated transport of urate in the 
terminal portion of the ileum. 

The authors gratefully acknowledge the secretarial 
assistance of Ms. Tofla M. Larkin. 

1. Weinman EJ, Sansom SC, Steplock DA, Sheth 
AV, Knight TF, Senekjian HO. The secretion of 
urate in the proximal convoluted tubule of the rat. 
Amer J Physiol 239 (Renal Fluid Electrolyte 
Physiol 8): F383-F387, 1980. 

2. Sansom SC, Senekjian HO. Knight TF, Babino H, 
Steplock DA, Weinman EJ. Determination of the 
apparent transport constants for urate absorption 
in the rat proximal tubule. Amer J Physiol 240 
(Renal Fluid Electrolyte Physiol 9):F406-F410, 
1981. 

3. Sorensen LB. The elimination of uric acid in man. 
Scand J Clin Lab Invest SuppI 54:1-214, 1960. 

4. Weinman EJ, Sansom SC, Steplock DA, Sheth AV, 
Knight TF, Senekjian HO. The use of high per- 



URATE TRANSPORT IN RAT SMALL INTESTINE 



23 



fbrmaiice bquid chroniatograpiiy for detennina- 
tioo of urate coocentratkNis in nanoliter quantities 
of fluid. Kidney Int 19:81-85, 1981. 

5. Kobssa N. Schutzenberser WG. Weiner H. Turn- 
hetm K. Active secretion of hypoiianthine and 
xanthine by guinea pig jejunum in- vitro. Amer J 
Physiol 238 (Gastrointest Liver Physiol 1):G141- 
G149, 1980. 

6. Berlin RD. Hawkins RA. Secretion of purines by 
the small intestine: Transport mechanism. Amer J 
Physiol 215:942-950. 1968. 

7. Scharrer E. Raab W, Tiemeyer W. Amann B. Ac- 
tive absorption of hypoxanthine by lamb jejunum 
in-vitro. Pflugers Arch 391:41-43, 1981. 

8. Hanisch ME, Lawrence AL. Purine and pyrimi- 
dine absorption by the gut of the chiton, cryp* 
tochiton Stelleri. Comp Biochem Physiol 
42A.-601-610, 1975. 

9. Shanker LS. Jeffrey JL, Tocco DJ. Interaction of 



purines with the pyrimidine transport process of 
the smaD intestine. Biochem Pharmacol 12:1047- 
1053, 1963. 

10. Wilson DW, Wilson HCW. Studies in-vitro of the 
digestion and absorption of purine ribonucleotides 
by the intestine. J Biol Chem 237:1643-1647, 
1962. 

11. Kahn AM, WUson S, CrawhaU J. The influx of 
uric acid and other purines into overted jejunal 
sacs of the rat and hamster. Canad J Pharmacol 
53:113-119, 1975. 

12. Oh JH, Dossetor JB, Beck IT. Kinetics of uric 
acid transport and its production in rat small in- 
testine. Canad J Physiol Pharmacol 45:121-127, 
1%7. 



Received February 8, 1982. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 24-33 (1982) 



Phosphoenolypyruvate (PEP) Effects on Fresh and Stored Red Blood 

Cells'* (41472) 

PAUL R. SOHMER* and RHONDA L. SCOTT+ 

* Division of Blood Research, and ^Division of Combat Casualty Care, Letterman Army Institute of 
Research, Presido of San Francisco, California 94129 



Abstract. The present studies were performed to evaluate the effects of phosphoenol- 
pyruvate (PEP) on fresh and stored red blood cells as a possible adjunct to citrate phosphate 
dextrose (CPD) for preservation of blood. Red cell concentrates prepared from fresh blood 
drawn from human volunteers were incubated with CPD alone or CPD containing PEP. At 
37° and an initial pH of less than 6.2, a significant increase in 2,3-DPG and maintenance of 
ATP were observed. A dose-dependent increase in 2,3-DPG was noted in blood incubated in 
CPD-PEP from 13 to 52 mM. 2,3-DPG did not change during incubation at either 4** or 25**. 
A significant increase in ATP was observed at 25°; ATP remained unchanged at 4° and 37°. 
Storage related depletion of 2,3-DPG and ATP was reversed by PEP incubation (26 mM) 
even when blood was stored for 42 days in CPD. In fact, 2,3-DPG levels two or three times 
greater than normal were regularly observed. The persistence, in vivo, of the PEP-induced 
increase in 2,3-DPG and Pso was demonstrated by 33% exchange transfusions in rats with 
homologous blood treated for 4 hr with 26 mA/ PEP. Unlike other preservation additives, 
incubation with PEP results in a dramatic increase in 2,3-DPG without depletion of ATP, a 
property which is maintained in vivo. Evidence shows that PEP should be considered as a 
potential adjunct to conventional blood preservation systems. 



Tomoda et al, (1) and Hamasaki et al. 
(2-4) have demonstrated that the glycoly- 
tic intermediate phosphoenolpyruvate 
(PEP) accumulates inside human erythro- 
cytes which have been incubated in acid- 
ified sucrose and citrate solutions contain- 
ing PEP. Transport of PEP across the red 
cell membrane and its subsequent metabo- 
lism results in a significant increase in the 
intracellular concentration of 2,3-diphos- 
phoglycerate (2,3-DPG) and adenosine tri- 
phosphate (ATP) during incubation in these 
media (1-4). 

The present studies were designed to as- 
sess the utility of PEP as an adjunct to (a 
presently available blood preservation 
system) citrate phosphate dextrose (CPD). 



' Presented in part at the American Federation of 
Clinical Research, Western Section Meeting, February 
5, 1981, Carmel. Calif. 

^ The opinions or assertions contained herein are 
private views of the authors(s) and are not to be con- 
strued as official or as reflecting the views of the De- 
partment of the Army or the Department of Defense 
(AR360-5). 



The importance of 2,3-DPG and ATP as in 
vitro indicators of erythrocyte function and 
viability suggests that a treatment which re- 
sults in their simultaneous generation or 
maintenance would significantly improve 
our ability to store blood in the liquid state. 
We have evaluated the effects of varying 
concentrations of PEP on both fresh and 
stored red blood cells. Time, temperature, 
and pH of incubation were varied to deter- 
mine optimum conditions for maximal gen- 
eration of 2,3-DPG and maintenance of ATP. 
Materials and Methods. Fresh blood. 
Fresh blood was drawn from healthy vol- 
unteers into syringes which had been pre- 
filled with standard citrate phosphate dex- 
trose (CPD) solution to achieve a final 
blood -anticoagulant mixture ratio of 7:1.^''* 



* Human subjects participated in these studies after 
giving their free and informed voluntary consent. In- 
vestigators adhered to AR 70-25 and USAMRDC Reg 
50-25 on the use of volunteers in research. 

* The citation of trade names in this report (or paper) 
does not constitute an official endorsement or ap- 
proval of the use of such items. 



24 
0037-9727/82/090024- 10$01 .00/0 

Copynghl 'C; 1982 by the Society for Experimental Biology and Medicine. 
**' -^"^fs reserved. 



PEP EFFECTS ON RED BLOOD CELLS 



25 



Red cell concentrates (RCC) were prepared 
after centrifugation in a Sorvall RC-3 refrig- 
erated centrifuge at 5000g for 5 min at 4°. 
Aliquots (4 ml) of the resultant RCC were 
incubated at yi" for 5 min in a shaking water 
bath (rpm = 140). Equal volumes of test 
solutions were added and samples removed 
for measurement of hemoglobin, 2,3-DPG, 
ATP, pH, and P50. The aliquots were then 
incubated at 4°, 25°, or 37° for 4 hr. Samples 
were removed at intervals for the bio- 
chemical analyses noted above. Test solu- 
tions were prepared by dissolving phos- 
phoenolpyruvate (tricyclohexylammonium 
salt obtained from Sigma Chemical Co., St. 
Louis, Mo.) in sterile CPD; the pH was ad- 
justed to 5.7 with dropwise addition of 0.1 
Af NaOHorO.lAf HCl. 

The first experiment was designed to 
evaluate the effects of the concentration of 
PEP on P50 and intraerythrocytic 2,3-DPG 
and ATP. Test solutions were prepared to 
achieve final concentrations (test solution 
plus RCC) of 0, 13, 26, 52, and 78 mAf PEP. 
After mixing and initial sampling, the RCC 
test solution mixtures were incubated at 37° 
for 4 hr. Samples for analysis of P50, 2,3- 
DPG, and ATP were removed after 1 and 4 
hr of incubation. 

The second experiment was designed to 
evaluate the effects of incubation tempera- 
ture of P50 and intraerythrocytic 2,3-DPG 
and ATP. Equal volumes of CPD or 52 mAf 
PEP in CPD were added to red cell con- 
centrates. The PEP and CPD-RCC mix- 
tures were incubated at 4°, 25°, and 37° for 4 
hr of incubation. 

The third experiment was designed to 
evaluate the effects of pH on intraeryth- 
rocytic 2,3-DPG and ATP. Equal volumes 
of CPD or 52 mAf PEP in CPD solution 
were added to red cell concentrates. The 
pH of the RCC solution mixtures was ad- 
justed to 6.2, 6.6, and 7.2 (pH measures at 
37°) by dropwise addition of 1 Af NaOH or 
0.1 Af HCl. The RCC solution mixtures 
were incubated at 37° for 4 hr. Samples 
were removed for biochemical analysis 
after 1 and 4 hr of incubation. 

In the fourth experiment, rats were sub- 
jected to a 33% exchange transfusion with 
either PEP-treated or control rat red blood 



cells to determine the in vivo persistence of 
the high 2,3-DPG and P50 induced by PEP 
incubation. Fresh rat blood was collected in 
CPD (1:7, V:V) from 16 retired breeder 
rats, 600-800 g. The donor rats were 
anesthetized with phenobarbital 50 mg/kg 
(Nembutal, Sodium, Abbott Lab., Chicago, 
111.) and bled via the abdominal aorta. 
Blood from two rats was pooled for each 
transfusion. Four pools were incubated 
with PEP for 4 hr as previously described 
(5) and four pools were held in the re- 
frigerator at 4° for 4 hr. Following incuba- 
tion or refrigeration, the red cells were 
washed twice in three vol of acid saline and 
resuspended in warm 4% albumin in 0.9% 
saline solution. The pH of the PEP-treated 
suspensions was 7.369 ± 002, the P50 was 
56.6 ± 4.7, and the 2.3-DPG content was 
31.45 ± 3.18 /Ltm/g Hgb. The pH of the con- 
trol suspension was 7.380 ± 0.011, the P50 
was 38.6 ± 0.7, and the 2,3-DPG content 
was 2.324 ±1.19 /itm/g Hgb. Male Sprague- 
Dawley rats, 250-300 g, were used as re- 
cipients. The animals were anesthetized as 
above, and the right exterior carotid artery 
and jugular vein were catheterized with PE- 
50 polyethelene and silastic (0.04 in 1.0) 
catheters, respectively. One-third of the 
total blood volume was exchanged by with- 
drawal of 2 ml/100 g body weight through 
the carotid cannula and replacement with 
an equal volume of cell suspension through 
the jugular catheter. Following transfusion, 
the jugular catheter was removed, and the 
carotid cannula was tunnled subcutaneously 
to the back of the neck and exteriorized be- 
tween the scapulae. The animals were then 
allowed to recover, and post-transfusion 
samples were drawn through the carotid 
cannula 6 hr following the transfusion. 

Stored blood, A total of 450 ml of blood 
was drawn from each of three healthy vol- 
unteers and collected into McGaw Haemo- 
Pak units containing 63 ml of CPD. These 
units were stored as whole blood (Hct = 
40 ± 5) under standard blood bank con- 
ditions at 4**. Aliquots (10 ml) were removed 
at weekly intervals and placed on ice. Base- 
line P50, ATP, and 2,3-DPG levels were 
measured. The remaining sample from each 
unit was divided equally and incubated at 



26 



PEP EFFECTS ON RED BLOOD CELLS 



37°, pH 5.9-6.2, for 4 hr in equivalent vol- 
umes of CPD or PEP in CPD. The final con- 
centrations of PEP in the experimental 
blood -solution mixtures was 26 mAf. The 
P50, ATP, and 2,3-DPG levels were mea- 
sured immediately after mixing and after 1 
and 4 hr of incubation in a shaking 37° water 
bath (rpm = 140) with either control (CPD) 
or experimental (PEP in CPD) solution. 

Analytical procedures. Pjowas measured 
by the biotonometry method described by 
Neville (5). Concentrations of ATP and 
2,3-DPG were determined from protein-free 
filtrates prepared with 6% perchloric acid 
(0.5:2.0, blood:PCA) neutralized with 
potassium bicarbonate. Assays were per- 
formed using commercially available as- 
say kits (ATP, Sigma technical bulletin 
366-UV; 2,3-DPG, Sigma technical bulletin 
35-UV). Glucose was measured in the 
protein-free filtrates using a hexokinase 
technique (7). 

Results. Fresh blood. The effects of vary- 
ing the concentration of PEP on intraeryth- 
rocytic 2,3-DPG are presented in Fig. 1. A 




Fig. I. Effect of varying PEP concentration on red 
cell 2.3-DPG production. Each value represents mean 
± SEM (// = 3) symbols without vertical bars indicate 
the standard errors were smaller than the symbol used. 
V, 78 mM: O, 52 mM: ♦, 26 mM; D. 13 mM; •. mM. 



progressive fall in 2,3-DPG occurred in 
CPD alone such that levels 25% of the ini- 
tial value were observed after incubation 
for 4 hr. Increasing the concentration of PEP 
from 13 to 52 mAf resulted in increased gen- 
eration of 2,3-DPG. Postincubation 2,3- 
DPG levels obtained with 78 mAf PEP did 
not differ from those obtained with a 52 mAf 
solution. The effect of varying the concen- 
tration of PEP on pH, P50, and ATP are 
presented in Table I. A progressive fall in 
ATP occurred in CPD alone; an increment 
in ATP was observed with all concentra- 
tions of PEP. Increments in P50 paralleled 
the rise in 2,3-DPG. An initial increase in 
P50 was observed following addition of CPD 
alone. This could be attributed to an al- 
kaline Bohr effect due to the low pH of the 
incubation mixture (8). However, the P^o 
increments observed with PEP incubation 
were significantly greater {P < 0.05) than 
those achieved with CPD alone, although 
the pH was the same in both CPD and PEP 
incubations. A moderate degree of hemoly- 
sis was observed when PEP was added to 
exceed a final concentration of 26 mAf. 

The effect of incubation temperature on 
intraerythrocytic 2,3-DPG is presented in 
Fig. 2. A 26 mM concentration of PEP in 
CPD was chosen because of hemolysis ob- 
served with higher concentrations. A dra- 
matic increase in 2,3-DPG was observed at 
37°. When measured at 4 hr, 2,3-DPG was 
three times greater than initial value. 2,3- 
DPG was maintained without significant 
change at 4° in both PEP- CPD and CPD 
alone. When the blood -PEP in CPD mix- 
ture was incubated at 25°, 2,3-DPG was also 
maintained at a level which approximated 
the initial value, whereas incubation at 25° 
with CPD alone resulted in a slight reduc- 
tion in 2,3-DPG concentration. The effects 
of incubation temperature on ATP are pre- 
sented in Fig. 3. ATP was maintained close 
to its initial value in PEP in CPD (and CPD 
alone) at 4° after incubation for 4 hr. How- 
ever, a small but significant increase (P < 
0.05) in ATP was observed at 25°. ATP was 
maintained close to its initial value at 37**. 

The effect incubation pH on intraeryth- 
rocytic 2,3-DPG and ATP is presented in 
Table II. A fall in 2,3-DPG concentration 



PEP EFFECTS ON RED BLOOD CELLS 



27 



u 
a 

& 

z 
o 
z 
o 



z 

u 

z 

5 

o 

z 



< 

> 



u 

u. 
u. 



ffi 



s 


o ^ 


+1 


+1 +1 


s 

'^ 


g5 



® 


s 


s 


+1 


+1 


+1 


s 


1 


3 


■<• 


» 



00 ri o 



+1 


+1 


+1 




1 


?; 


-^ 




00 




>£> m 



•o 


o So 


+1 


+1 +1 




\0 r*^ 



+1 


+1 


+1 




s 





vo 


ss 


+1 


+1 +1 







00 



o 
6. 






4> 



•e3 
•a 
S 



^ !r < 



O Q 

o ri 

+1 +1 +1 



g S2 



+1 +1 

O On 

2^ 



+1 +1 +1 



+1 +1 +1 

00 r^ On 





i.h 


+1 


+1 +1 


00 





^ O >o 

+1 +1 +1 



00 



•^ i 



3C 8 



S:5 



28 



PEP EFFECTS ON RED BLOOD CELLS 




4* c »•• c »y • c »• • c 



CTO^ff^ 



CPO 



Fig. 2. Effect of temperature of incubation on red 
cell 2>DPG. At each temperature tested, the left bar 
reprenems initial mean 2.3-DPG; the right bar repre- 
sents 2.3'DPG measured after 4 hr of incubation. Ver- 
tical bars represent SEM (/i = 4). A paired / test was 
used to determine if the 2,3-DPG concentrations were 
different before and after incubation at each tempera- 
ture. There was a significant (P < 0.001) increase when 
cells were incubated at 37" in the presence of PEP. 
There was a significant iP < 0.001) decrease when 
cells were incubated at 25"" in CPD akme. 



which was similar to that observed with 
CPD alone occurred at both pH 6.7 and 7.2 
in red cells which had been incubated in 
PEP in CPD. However, incubation at pH 
6.2 resulted in a 100% increase in 2,3-DPG. 
ATP concentration increased or was main- 
tained regardless of the pH of incubation 
with PEP in CPD whereas, with CPD alone 
it was maintained at pH 6.7 and 7.2, but 
decreased at pH 6.2. 

Stored hlood. The effect of storage and 
subsequent incubation with either CPD or 
PHP in CPD on P.^, is presented in Table III. 
As expected. P..,,, decrease progressively 
relative to the length of time in storage in 
CPD from Day 14 to Day 42. On all days 
studied, post-storage incubation with CPD 
alone resulted in an initial increase in P50 
which can be attributed to the reduced pH 
of the incubation medium and an alkaline 
Bohr effect. Incubation with PEP in CPD 
resulted in P5,, values which were signifi- 
cantly greater than that achieved with CPD 
alone. This effect was noted on all days 
studied. 

The effect of storage and subsequent in- 




trc 



3r c 



tfc 



CPD 4- PEP CPO 

Fig. 3. Effect of temperature of incubation on red 
cell ATP. At each temperature examined, the left bar 
represents initial mean ATP, the right bar represents 
ATP after 4 hr of incubation. Vertical bars represent 
the SEM in = 4). A paired / list was used to determine 
if the ATP concentration after incubation differed from 
the initial value. There was a significant increase (P < 
0.05) in ATP concentration only when cells were incu- 
bated in the presence of PEP at 25°. 



cubation with either CPD or PEP in CPD on 
red cell 2,3-DPG is presented in Table IV. 
As expected, red cell 2,3-DPG decreased 
progressively during storage in CPD. Post- 
storage incubation with CPD alone resulted 
in a further decrement in 2,3-DPG on all 
days studied. However, incubation with 
PEP in CPD resulted in a significant in- 
crease in 2,3-DPG; levels two to three times 
greater than those found in fresh red blood 
cells before storage were observed even 
after 42 days of />? vitro storage. 

The effect of storage and subsequent in- 
cubation with either CPD or PEP in CPD on 
red cell ATP is presented in Table V. Post- 
storage incubation with CPD alone resulted 
in a reduction in intraerythrocytic ATP. 



PEP EFFECTS ON RED BLOOD CELLS 



29 



X 



5 

OB 

u 
z 

2 

U 

o 



■+ 
D 
U 



^ ® ® 






o 
+1 



o 
+1 



o 






00 f*i 



00 o 

d d 

+1 +1 



8 I 



ffi 

Q 

& 

Z 

o 

z 
o 

n 

D 
U 

z 



o 
X 



u. 
O 

u 



ffi 
< 



K +1 



o 
+1 



D 
Pu. 

U 






«n fo 



S 

bu 



+1 



o 
+1 

s 



s 

d 

+1 

8 



+1 
8 






o 






-§ 






3 

E 

§ 

c 

3 

Q 



J3 
00 



■a 



p {« 

^ ed >. 

^ > S 

3 O o 
•e3.S2 «= 
>(/)«) 

o.w< a. 

— *• « *C 
o 



However, incubation with PEP in CPD re- 
sulted in a slight increment or maintenance 
of ATP on all days studied. 

The in vivo effects of 33% exchange 
transfusion with PEP treated or control rat 
red cells are presented in Table VI. Al- 
though only one-third of the blood was re- 
placed, there was a significant increase in 
P50 following transfusion. This result indi- 
cates (a) the cells are viable and are not 
cleared by 6 hr post-transfusion and (b) that 
there is no immediate catabolism of the su- 
pranormal 2,3-DPG levels following trans- 
fusion. 

Discussion. Since Chanutin and Cumish 
(9) and Benesch and Benesch (10) first ob- 
served that the interaction of 2,3-DPG with 
hemoglobin reduces hemoglobin-oxygen 
affinity, the importance of 2,3-DPG as a 
mediator of oxygen delivery has been con- 
firmed both in vitro and in vivo (11). A 
functional red cell defect reflected by an in- 
crease in hemoglobin-oxygen affinity has 
been directly correlated with depletion of 
2,3-DPG in all conventional liquid blood 
preservation systems. A variety of addi- 
tives have been employed to maintain red 
cell 2,3-DPG and hemoglobin function 
during prolonged liquid blood storage. In 
general, these additives have stimulated in- 
creased generation of 2,3-DPG at the ex- 
pense of the ATP (12-14) whose contribu- 
tion to the maintenance of red cell mem- 
brane integrity (15, 16), deformability (17), 
and post-transfusion viability (18, 19) is 
well accepted. Thus, hemoglobin function 
may be preserved at the expense of red cell 
viability. Conversely, additives such as 
adenine, which stimulate generation of ATP 
do so at the expense of 2,3-DPG (20); thus, 
red cell viability may be preserved at the 
expense of function. 

In the present studies, we have con- 
firmed the earlier reports by Tomoda et al. 
(1) and Hamasaki et al, (2-4), regarding 
incubation of human red blood cells in a 
citrate solution which contains PEP. In ad- 
dition, we have demonstrated that the in- 
cubation of red blood cells stored for long 
periods in CPD with PEP results in the gen- 
eration of 2,3-DPG and the simultaneous gen- 
eration of ATP. Unlike other preservation 
additives, PEP may facilitate the mainte- 



30 



PEP EFFECTS ON RED BLOOD CELLS 



TABLE IIL Eff'WT of Incltbation with PEP on P»of Red Cells Stored in CPD 





Initial 




Ps* 


(mm Hg> 








I hr'^ 




4lir'- 


r>ay 


CPD 


PEP 


CPD 


PEP 





25,52 ± 1.03 


33.24 ± 2,41 


40.95 ± 2.49 


34.16*3.44 


48.63* ± 1.43 


7 


24.45 ± 3.26 


33.36 ± 3.34 


39.67 ± 2.56 


35.64 - 3.8 


48.74* ± 2.31 


14 


20.31 ^ 2.30 


28.39 ± 1.67 


39.44* ± 1.40 


33.80-12 


49.05* ± .73 


21 


17.85 ± .65 


26.04 ± I.IO 


38.96* ± .75 


34.01 ±2.18 


52.25* ± 1.53 


28 


18.07 t .36 


25.74 ± .04 


39.82* ± .62 


34.47 ± .52 


56.77* ± 2.67 


35 


18.75 ^ .71 


28.00 i .98 


41.22* ± 1.29 


36.82 ±2.19 


60.62* ± 2.67 


42 


17.89 ± 1.16 


24.37 ± 1.35 


38.52* ± 2.84 


29.90 ± 2.05 


53.03* ± 6.66 



' finch value repre%eni% mean ± SEM; /i = 3. 

' On each sampling clay. compariM>n» were made between cells incubated in CPD alone and CPD + PEP at 
each time period; an aMerink (*) indicates a significant difference was found using a paired / test iP < 0.05). 



nance of both hemoglobin function and red 
cell viability during or after (rejuvenation) 
prolonged blood storage. However, the re- 
sult n of our efforts to optimize the PEP 
treatment clearly demonstrate that its ef- 
fectiveness is dependent on the mainte- 
nance of incubation conditions which are 
generally considered to be detrimental to 
human red blood cells. 

Tomoda ct al, (1) and Hamasaki et aL 
(2 4) have demonstrated that PEP ac- 
cumulates inside human erythrocytes which 
have been incubated in acidified sucrose 
and citrate solutions (pH = 4.5-6.5). Trans- 
port is apparently mediated via a mem- 
brane-bound anion transport system. In the 
presence of solutes to which the red cell 
membrane is impermeable (sucrose and 



citrate) the Donnan equilibrium is main- 
tained such that PEP can accumulate within 
the red cell against the concentration gra- 
dient (2). The present studies confirm the 
pH dependency of the PEP transport. We 
have observed that the optimum pH range 
is 5.9 to 6.2 using standard CPD solution as 
the PEP solvent. Above this range, PEP has 
no effect on incubated red blood cells. 

The effect of PEP on P50 is negligible at 
4**. This suggests that the action of PEP is 
mediated by an enzymatic mechanism since 
at temperatures which are closer to the op- 
timum for glycolytic activity a dramatic in- 
crement in P50 is observed. In our work 
with species other than human (21) we have 
observed that the reduction in hemoglo- 
bin-oxygen affinity associated with PEP 



lAHIl' IV. 1:1 IK I 01 Incubaiion with PEP on 2. 3-DPG Content of Red Cells Stored in CPD 





tniliar^ 
10.42 • I.IV' 


1 
8.(H) 




2,3-DPG (Mmol/g Hgb) 






Post-CPC 


y.2 

4hr 


Post-PEP»^^ 


Day 


hr 


1 hr 


4hr 





♦ 1.32"'' 


3.90 ± 2.38" 


24.97 ± 4.57'- 


42.85 ± .15" 


7 


10.07 • l.W 


8.47 


*. 3.52"'' 


5.23 ± 1.99" 


20.83 ± 3.29' 


48.13 ± 6.93" 


14 


1.22 ' 0.84" 


3,50 


.♦. I.3I" 


2.87 ± 0.94" 


16.00 ± 3.16" 


41.70 ± 1.56'- 


:i 


2.48 ♦ l.(>ft" 


l.f>7 


± 0.44" 


0.63 ± 0.38" 


16.50 ± 0.47" 


38.30 ± 1.97^ 


28 


0.84 • O.V^" 


0.50 


* 0.85" 


0.37 ±0.12" 


14.80 ± 0.75" 


33.10 ± 1.97'' 


\s 


22 ♦ 0.10" 


0.(H) 


♦ 0.00" 


0.00 ± 0.00" 


12.07 ± 0.20" 


33.97 ± 1.39'- 


41 


<3 • 0.20" 


0.83 


♦ 0.42" 


0.10 ± 0.12" 


11.53 ± 0.54" 


31.10 ± 3.07'- 



' lach value toptCNcnls mean * SV.M: n 3. 

* On each sampling ilay. analysis of variance was performed initially; Duncan's multiple range test was then 
pcrfonncil to ilclcrminc which means varied. Means within rows with same superscripts arc not significantly 
ilifTcivnt. 



PEP EFFECTS ON RED BLOOD CELLS 



31 



TABLE V. Effect Incubation with PEP on ATP of Red Cells Stored in CPD 





Initial*-* 




ATP Oimol/g Hgb) 






Post-CPD»-* 


Post-PEF* 


Day 


Ihr 


4hr 


1 hr 


4hr 





4.76 ± 0.25* 


4.47 ± 0.30* 


3.53 ± 0.24° 


5.60 ± 0.49* 


4.80 ± .33* 


7 


4.25 ±0.16'' 


3.20 ± 0.44* 


2.17 ±0.12" 


4.23 ± 0.09^ 


3.67 ± 0.64*'^ 


14 


3.77 ± 0.38*'' 


3.00 ± 0.35* 


1.57±0.18« 


3.90 ± 0.27'^ 


3.80 ± 0.25*-^ 


21 


2.70 ± 0.25*'' 


2.03 ± 0.36* 


1.23 ±0.20° 


3.60 ± 0.30" 


3.13 ±0.23^ 


28 


2.28 ±0.19* 


1.83 ±0.24°* 


1.20 ±0.15° 


3.30 ± 0.40'^ 


3.17 ±0.23'- 


35 


2.32 ± 0.30''* 


1.87 ±0.18'' 


1.43 ±0.39° 


3.73 ± 0.43*^ 


3.20 ± 0.36*-'^ 


42 


1.80 ±0.12* 


1.57 ±0.19* 


0.97 ± .09° 


2.87 ±0.15^ 


2.67 ± 0.23'- 



* Each value represents mean ± SEM; /i = 3. 

* On each sampling day, analysis of variance was performed initially; Duncan's multiple range test was then 
performed to determine which means varied. Means within rows with same superscripts are not significantly 
different. 



incubation is limited to those species whose 
hemoglobin is sensitive to 2,3-DPG, al- 
though in some cases, the PEP effect is 
limited by species specific differences in 
membrane transport. Furthermore, PEP 
fails to react directly with stroma-free he- 
moglobin (P. J. Scannon, unpublished data). 
At the pH of incubation required for PEP 
transport, it is unlikely that 2,3-DPG is gen- 
erated by the upper portion of the glycolytic 
pathway since phosphofructokinase, the rate- 
limiting enzyme, is certainly inhibited. We 
observed no difference in final glucose con- 
centration following incubation in CPD or 
CPD plus PEP (data not shown). The 
mechanism of accumulation of 2,3-DPG ob- 
served in red blood cells which have been 
treated with PEP may be comparable to 
that observed in pyruvate kinase deficiency 
where a defect occurring late in glycolytic 
pathway results in an abnormal accumula- 



tion of PEP, 3-phosphoglycerate, and 2,3- 
DPG (22). In patients who suffer from this 
hereditary abnormality of red cell gly- 
colysis, 2,3-DPG levels may be two to three 
times greater than normal and P50 may be 
significantly elevated (23, 24). The coin- 
cident maintenance of ATP which was ob- 
served in the present studies, however, 
suggests that pyruvate kinase function per- 
sists and that some PEP is dephosphory- 
lated to produce pyruvate. This is con- 
firmed by the attendant increments of py- 
ruvate noted by Tomoda et al, (1) in their 
initial experiments. 

PEP is a high-energy intermediate of the 
glycolytic pathway. As an intermediate 
compound, it is usually rapidly dephos- 
phorylated in the presence of magnesium 
and potassium ions to form pyruvate. This 
reaction is catalyzed by pyruvate kinase 
and results in production of ATP. This is 



TABLE VI. Effect of Exchange Transfusion on In Vivo 2,3-DPG P50 in Rats 



Control*" 



PEP-treated** 



Pretransfiision 2.3-DPG (/xm/g hgb) 21.25 ± 1.60 

Pretransfiision P50 (Torr) 39.64 ± 1.24 

Post-transfusion 2,3-DPG (/xm/g Hgb) 21.81 ± 1.34 

Post-transfiision P50 (Torr) 39.28 ± 2.55 

Post-transfusion pH 7.39 ± .03 



22.04 ± 0.87 

37.41 ± 1.03 

27.58 ± 1.44* 

46.0 ± 2.37* 

7.37 ± .02 



* Each value represents mean ± SEM of samples from four rats. 

^ Paired / test was used to compare pre- and post-transfusion 2,3-DPG content and Pso of blood from trans- 
fused rats. Asterisk {*) indicates a significant difference {P < 0.05). 






i--. « .•^'»*«-^. I»^' "^' li 



♦ r r . 



■ ■*' Jtr ' ■y 









/^ 






y^ ^./ ..... ^,- •:^.- . . -- - - . . 1-14- "::r?:ii,:r 






-■- C TufTi- 



'y 



• X ^ v* ' ^ 



,' , • - § ■ * 



s ... . ^ 



/ 



< ,, / . . Iif/.ii' 'I '. / 'f.' 
, , I ' . I .•■ iM'l H ' ith 

I . r r . Kilt ■ ,>li' r I'f ' ' I /.I 

II. .1 i«. ■<. • i»h ri r »• »iu . 

r Ml ' ' I 'T' I 'ithniil 



; r- .' •■ • .-.r-. - ♦.* -.-=.■ ■-^'^iw Jill 

■■',..:■■. :-. -rr,;." : ^ r.xz ^r r'-L'.iiiJr-urr 
• -.r-.i:- -. ■ -..■'. --.- i » -Tc-^ 5;i:cr" - 

'■■'••/■'* --■ i ^' .: : -■■•■:- .-.1- -----.. -e ; - -J^e 

';" r ,■ , ' -.-.',• ■ '. "•;".*. ■ ~ B i:v;":o"r 

f ,•..•.", / ' . '■ .- I/i -1 •^" 

i'^' .#s« • - r'r: ".'-:■-." -"-i'we of 
' • ;*('^ ; ■■..■; .; ■-:r .•^... -, r:;- Lab 

. . 1 1 ,'l .'/ > 
I ' I'., .jr. JM If ,r,'{..' :♦■■: f H P'j-.cr.a'i.T. of red 
ki#.','! « II ./ ', iliphii .f,hiiij.'i..cr.i-c iM ^lored hlooU 
• i.nr .i/iif,jf ilili ,i|f'i' ,;i' itIiidc fr;ln^fu^Ion 13:84, 



( , I ■ . . I II ' I. I I' Mi'.H ll th l\ j,^^, 

'• ' ■• ' i.m.l/M.jM Ill ,, ^, , , |j.,„|.,l Ihr cHcti of ascorbale on 

■ I. •• I I II 'I I I ill' ii. ..II. . Mil. ,,,,, I ' UJi'ri III sioifil icilcclls, Brit J 

If. ,. Ijfl n I. |illi< •! I.ii ri I' -|. II .. Ill ih.l .". M I Mi/i 

II . f f I , I t iii.iMi'il I .iMili II M.iiiii (il {•iljiiiilMI- I xti-iuU'ti N(H>d Storage 



PEP EFFECTS ON RED BLOOD CELLS 



33 



using optional additive systems containing 
ascorbate-2-phosphate. Submitted for publica- 
tion. 

15. Weed RT, LaCelle PC. ATP dependence of eryth- 
rocyte membrane deformability. In Green wait TJ. 
(Ed.), Red Cell Membrane Structure and Func- 
tion, Philadelphia, Lippincott, 1969, p. 318. 

16. Weed RT, LaCeUe PC, MerriU EW. Metabolic de- 
pendence of red cell deformability. J Clin Invest 
48:795, 1%9. 

17. LaCelle PC. Alterations of deformability of the 
erythrocyte in stored blood. Transfusion 9:238, 
1969. 

18. Nakao K, Wada T, Kamiyama T. A direct re- 
lationship between adenosine triphosphate and in 
vivo viability of erythrocytes. Nature (London) 
194:877, 1%2. 

19. Dem RJ, Brewer GJ, Wirokowski JJ. Studies on 
the preservation of human blood. II. The relation- 
ship of erythrocyte adenosine triphosphate levels 
and other in vivo measures to red cell storageabil- 
ity. J Lab Clin Med 69:968, 1967. 

20. Sugita Y, Simon ER. The mechanism of action of 



adenine in red cell preservation. J Clin Invest 
44:629, 1965. 

21 . Scott RL, Sohmer PR. Comparative aspects of the 
effect of phosphoenolpyruvate on mammalian 
erythrocyte metabolism. Clin Res 29:42A, 1981. 

22. Minakami S, Yoshikawa H. Studies on erythro- 
cyte glycolysis. II. Free energy charges and rate 
limiting steps in erythrocyte glycolysis. J Biochem 
59:139, 1966. 

23. Tanaka KR, Paglia DE. Pyruvate kinase defi- 
ciency. Semin Hematol 8:367, 1966. 

24. Oski FA, Marshall BA, Cohem PJ. Exercise with 
anemia: The role of the left shifted or right shifted 
oxygen hemoglobin equilibrium curve. Ann Intern 
Med 74:44, 1971. 

25. Hasart E, Jacobasch G, Rapoport S. Der Einfluss 
der Temperature auf die Regulation der glycolyse 
in Erthrozyten des Menchen und der Ratte. Acta 
Biol Med German 24:725, 1970. 



Received September 2, 1981. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 34-40 (1982) 



Effects of Sodium Chloride on Prostacyclin-Stimulated Renin Release in Dogs 
with Filtering and Nonfiltering Kidneys (41473) 

DANIEL VILLARREAL, RONALD H. FREEMAN, JAMES O. DAVIS,* 
JOHN R. DIETZ, and STEPHEN F. ECHTENKAMP 

Department of Physiology, University of Missouri School of Medicine, Columbia, Missouri 65212 



Abstract. The effects of intrarenal infusion of sodium chloride (NaCl) on prostacyclin 
(PGIs)-stimulated hyperreninemia were examined in groups of anesthetized dogs with either 
a single filtering kidney or a single denervated nonfiltering kidney, a model in which the renal 
tubules are damaged, and the macula densa is nonfunctional. After control observations, 
intrarenal infusion of prostacyclin at nonhypotensive doses resulted in significant incre- 
ments of renin secretion and renal blood flow (RBF) in both preparations. Superimposition 
of intrarenal NaCl to the ongoing prostacyclin infusion produced a striking decrease of renin 
secretion in dogs with a filtering kidney. In contrast, dogs with a nonfiltering kidney failed to 
show a significant change in renin secretion during intrarenal NaCl administration. Renal 
blood flow remained unaffected by NaCl in both groups. The increment in renal venous 
plasma sodium concentration of 18-21 meq/liter was similar in both series. It is proposed 
that the renin response to intrarenal NaCl was mediated through the renal tubules, since 
renin secretion failed to decrease in the nonfiltering kidney preparation. Thus, the present 
results indicate that prostacyclin-stimulated renin secretion was modulated by a tubular 
mechanism, probably the macula densa. 



The influence of the renal prostaglandins 
on renin secretion is well established. It has 
been demonstrated that intrarenal infusions 
of arachidonic acid and various prostaglan- 
dins, including PGI2, PGE2, and PGD2 can 
stimulate renin secretion in the dog (1-5). 
Of the several known renal prostaglandins, 
it has been suggested that prostacyclin may 
play a major physiological role in the re- 
lease of renin (4, 6). Indeed, Frolich et al. 
(7) have postulated that prostaglandin- 
dependent renin release is probably medi- 
ated by PGI2. Although the mechanism of 
prostacyclin stimulation of renin is not well 
defined, studies in vitro (8, 9) have demon- 
strated that PGI2 exerts a direct action on 
the juxtaglomerular (JG) cells. Prostacyclin 
stimulation of renin secretion also has been 
demonstrated in the nonflltering as well as 
the filtering kidney preparation of anes- 
thetized dogs (2, 4, 10). Therefore, it 
appears that a tubular or macula densa 
mechanism is not essential for the in vivo 
response to prostacyclin infusion. How- 



' To whom reprint requests should be addressed. 



ever, in these earlier studies (4, 10) during 
intrarenal prostacyclin administration into 
filtering kidneys, a marked natriuresis also 
occurred; it is possible that this effect could 
have attenuated the PGl2-dependent hyper- 
reninemic response. These considerations, 
combined with the abundant physiological 
data which support an important role for a 
tubular or macula densa mechanism in the 
control of renin secretion (11, 12), led us to 
design the present experiments to examine 
the potential influence of the renal tubular 
system on prostacyclin-stimulated hyper- 
reninemia. Thus, changes in PGl2-stim- 
ulated renin release were studied in either 
single filtering or denervated nonfiltering 
kidneys of dogs before and during the in- 
trarenal infusion of hypertonic NaCl. In a 
previous investigation. Shade et al. (13) 
demonstrated an inhibition of the hyperse- 
cretion of renin secondary to chronic caval 
constriction during intrarenal infusion of 
hypertonic NaCl into dogs with filtering 
kidneys, but not in dogs with nonfiltering 
kidneys. Therefore, the present experi- 
mental design is an approach to examine 
the importance of the renal tubules and pre- 



34 
0037-9727/82/090034-07$01 .00/0 

'"onyright © 1982 by the Society for Experimental Biology and Medicine. 
^^ts reserved. 



PIIOSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



55 



sumaMy the maciih densa in the control cf 
prostacyclin-induced hypenenmemia. 

Methods. The study involved 13 female 
mongrel dogs with body weights between 
13 and 22 kg. Prior to the acute experiment, 
the dogs were housed in individual metabo- 
lism cages and maintained on a diet that 
provided approximately 65 meq of sodium 
and 35 meq of potassium daily for at least 4 
days before the acute study and until 
sodium excretion approximated sodium 
intake. Water was available ad libit urn. 

All acute experiments were performed in 
the postabsorptive state. On the morning of 
the experiment, the dogs were anesthetized 
with sodium pentobarbital (30 mg/kg« iv), 
and supplemental small doses of the drug 
were administered as needed to maintain a 
satisfactory level of anesthesia. Pblyethyl- 
ene catheters (fir. 8) were inserted into the 
femoral vessels and their tips advanced to 
the inferior vena cava and abdominal aorta. 
The arterial catheter was connected via a 
Statham P23Db strain gauge pressure 
transducer to a Hewlett Packard 7702B 
recorder in order to monitor mean arterial 
pressure (MAP). The femoral vein catheter 
was used for the administration of cre- 
atinine and replacement of blood. Both 
kidneys were exposed via retroperitoneal 
flank incisions and a unilateral nephrec- 
tomy was performed to remove any influ- 
ence of this kidney on renin secretion. The 
contralateral renal artery was fitted with an 
electromagnetic flow probe connected to a 
model 501 Carolina Electronics flowmeter 
for determination of renal blood flow. 
Number 22-gauge needles attached to poly- 
ethylene tubes were inserted in the renal 
artery (distal to the flow probe) for infusion 
of solutions with a Harvard syringe pump 
and into the renal vein in order to obtain 
renal venous blood samples. A ureteral 
polyethylene 100 catheter was also posi- 
tioned near the renal pelvis and urine was 
collected in a graduated cylinder. After a 
priming solution of creatinine was given, a 
constant infusion was maintained through- 
out the experiment in order to determine 
creatinine clearance (Ccr)* In addition, an 
isotonic NaCl infusion into the renal artery 
was begun. All intrarenal solutions includ- 



ing prostacyclin, isotonic NaCK and hyper- 
tonic NaCI were given at a constant rate of 
0.59 ml/min. Prostacyclin sodium sah was 
prepared by dissolving 1 mg in 1 ml of 1 .1/ 
Tris buffer: this solution was diluted to 100 
ml with 0.05 M Tris buffer (pH 9.4). The 
final dilution for the desired concentration 
was made with isotonic NaCI. Concentra- 
tions of plasma arterial sodium, potassium, 
and creatinine were determined from bkxxi 
samples obtained at the midpoint of each 
period and in the urine collected over the 
entire clearance interval. Blood for the 
measurements of renal venous plasma 
sodium (RVP Na) was obtained at the end 
of each period. From these data* Co and 
electrolyte excretion were calculated. 
During the last 4 min of each period, renal 
venous and arterial blood samples were 
collected for measurement of plasma renin 
activity (PRA) and hematocrit. Mean arte- 
rial pressure and RBF were recorded con- 
tinuously. Folk>wing surgical preparation, a 
60-min equilibration period was alk>wed 
before initiation of the following two ex- 
perimental series. In all experiments, blood 
withdrawn for sampling was replaced with 
an equal volume of fresh donor bk>od from 
a normal dog. 

Experiment /. Intrarenal PGIt and su- 
perimposed NaCI infusion in anesthetized 
dogs with a single filtering kidney (N =7). 
Three 15-min control renal clearance peri- 
ods with intrarenal infusion of isotonic 
NaCI were obtained. In the subsequent two 
periods, prostacyclin was given intrarenally 
at a rate of 1.0 x 10~^ g/kg/min. This was 
followed by two periods of prostacyclin 
infusion at a rate of 1.5 x 10~^ g/kg/min. 
This latter dose of prostacyclin was con- 
tinued throughout the remainder of the ex- 
periment in order to achieve maximal 
stimulation of renin release without a con- 
comitant drop in MAP. During the next 
three periods, PGI2 and hypertonic NaCI 
were given together intrarenally. The 
hypertonic NaCI solution was administered 
in a concentration calculated to increase the 
existing RVP Na concentration an addi- 
tional 20 meq/liter for each individual ex- 
periment and the range for the group was 
2.42-5.84 meq/min. Finally, two recovery 



36 



PROSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



periods with infusion of PGI2 alone were 
obtained. 

Experiment 2, Intrarenal PGI2 and su- 
perimposed NaCl infusion in anesthetized 
dogs with a single denervated nonfiltering 
kidney (N = 6). A denervated nonfiltering 
kidney was produced in each dog 4 days 
before the acute experiment. The nonfil- 
tering kidneys were prepared according to 
the method of Blaine et aL (14). Briefly, 
under anesthesia with pentobarbital, one of 
the dog's kidneys was exposed via a ret- 
roperitoneal flank incision. The renal ar- 
tery was occluded for 2 hr with a serraflne 
clamp. After restoration of blood flow, the 
ureter was ligated and cut. Finally, the kid- 
ney was denervated. Four days later the 
acute experiment was performed and an 
identical protocol described for series No. 1 
was followed, except that due to the ab- 
sence of filtering capacity, ureteral cath- 
eterization was omitted; consequently, 
measurements of Ccr and electrolyte excre- 
tion were not obtained. Also in this series, 
the concentration of hypertonic NaCl infu- 
sion was adjusted to increase the existing 
RVP Na concentration an additional 20 
meq/liter for each individual experiment. 
Since the renal plasma flow in the nonfll- 
tering kidney was substantially reduced 
when compared to the filtering kidney, the 
rate of infusion of hypertonic NaCl was less 
than in the filtering kidney series, and it 
ranged from 1.6 to 2.5 meq/min. At the end 
of each experiment, these kidneys were de- 
capsulated and a solution of lissamine green 
dye was injected into the renal artery to 
verify the nonfiltering status. This was ac- 
complished by the failure to observe the 
appearance of dye within the renal tubules. 
This criterion was fulfilled in each of the 
dogs reported in this series. 

Analytical methods. Plasma and urine 
electrolyte and creatinine concentrations 
were determined by standard techniques. 
PRA was measured by radioimmunoassay 
for angiotensin I(A-I) with the technique 
described by Sealey et al., (15). Renin se- 
cretion rate was calculated by multiplying 
the difference between renal venous and 
arterial PRA concentration by renal plasma 
flow. Renal plasma flow (RPF) was calcu- 



lated from the formula RPF = RBF x 1 - 
HCT. Hematocrits were determined by the 
microcapillary tube method. 

The results are presented as an average 
of the control and each of the different ex- 
perimental categories. The data were ana- 
lyzed by the x* sign test for differences 
between categories. AP value of <0.05 was 
considered significant. 

Results. Experiment 1. Effects of intra- 
renal PGI2 and hypertonic NaCl infusion in 
anesthetized dogs with a single filtering 
kidney. Infusions of PGI2 at nonhypoten- 
sive doses of 1.0 and 1.5 x 10"® g/kg/min 
significantly increased renin secretion from 
147 ± 70 to 357 ± 146 and 416 ± 131 ng 
A-I/min, respectively (Fig. 1). Superimpo- 
sition of hypertonic NaCl during PGI2 ad- 
ministration resulted in an increase in RVP 
Na concentration of approximately 18 
meq/liter (P < 0.05) (Table I) and a 
threefold increase in renal sodium excretion 
{P < 0.05) (Table I). Also, the filtered load 
of sodium appeared to increase in 6 out of 7 
animals although the increment did not 
reach the 5% level of significance for the 
group. These changes were correlated with 
a significant reduction of renin secretion 
from 416 ± 131 to 157± 51 ng A-I/min (Fig. 
1), a value that was not different from con- 



«-P<.05 from C 
^ -P<.05 frtxn NaCl 



(N-7) 
NaCl 



RENAL 

BLOOD ^\ pi-^ 

FLOW ^ 

(mlAnin) 



RENIN 

SECRETION 

(ngAI/min) 



r4n rh 



PGIjOO^Ugfiw*} ^ 



r-; 



Fig. 1 . Responses to intrarenal PGI2 and NaCl infu- 
sions in dogs with a single filtering kidney. Values are 
expressed as means ± SE. 



PROSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



37 



a 

z 
< 



o 
z ^ 

Ik 

2 » 



i ^ 

O Z 

s < 

I o 

Z u] 
u z 
OC D 

|i2 
*i 

^£ 

> o 

O H 

o £ 

Si 

« s 

O — 
Z o 

:5 * 
= z 



QQ 
< 



c: 



o 

o 



c 



P ^ O "^ '^^ 

"2 +' +1 +' +1 +1 +1 +' +1 



9 •^ON '^^ 

;2 +' +1 +' +1 +1 +1 +1 +1 

2 S • • • "^ • 



"^©•o — onooo 

'2 +1 +1 +1 +1 +1 +1 +1 +1 
— » r^moo* » » 

cij § ^ :t c<i s S ^ 



5 +1 +1 +1 +1 +1 +1 +1 +1 
m» Qmoo* » » 

— -^ ^ f^ iTi w-^ »- 



U +1 +1 +1 +1 +1 +1 +1 +1 









ill « s-^fi 



MS 



0.0. 



11 

-I 
§.3 



er 



as 
1.1 



I 



. B 
58 
St, 
■E-S 
|8. 

•is . 

« «> O 

^ p I 
CO g.S 

lis 

S «« I 

(^^ i^ 

4> M O 
3'C C 
ed *> u 

^ i> 



trol. During the recovery period with infu- 
sion of PGI2 alone, renin secretion ap- 
peared to rise toward the prehypertonic 
NaCl level, but the increment was not sig- 
nificantly different from the control (Fig. 1); 
however, both RVP Na and urinary sodium 
excretion remained elevated during the re- 
covery (Table I). PGI2 infusion induced a 
significant increase in the renal blood flow 
which was not attenuated during hypertonic 
NaCl administration (Fig. 1). Likewise, 
renal vascular resistance (RVR) was de- 
creased with PGI2 infusion (both doses) (P 
< O.OS) and remained attenuated during the 
hypertonic NaCl infusion and recovery pe- 
riod (P < 0.05) (Table I). Creatinine clear- 
ance and MAP were not detectably changed 
throughout the experiment (Table I). PGI2 
significantly increased renal potassium 
excretion, but there were no further con- 
sistent changes in the excretion of this 
electrolyte during hypertonic NaCl admin- 
istration (Table I); during the recovery pe- 
riod, potassium excretion was again ele- 
vated (P < 0.05) (Table I). In addition, renal 
venous plasma potassium (RVP K) con- 
centration decreased slightly (P < 0.05) 
during hypertonic NaCl infusion and during 
the recovery period (P < 0.05) (Table I). 

Experiment 2, Response to intrarenal 
PGI2 and hypertonic NaCl infusion in 
anesthetized dogs with a single denervated 
nonfiltering kidney. As in experiment 1, 
infusion of nonhypotensive doses of PGI2 at 
a rate of 1.0 and 1.5 x 10"® g/kg/min re- 
sulted in a significant increase in renin se- 
cretion from 1 19 ± 30 to 456 ± 53 and 463 ± 
42 ng A-I/min, respectively (Fig. 2). Intra- 
renal infusion of hypertonic NaCl produced 
a significant increase in RVP Na concen- 
tration of approximately 20 meq/liter (Table 
II). In contrast to series 1, the superimposi- 
tion of hypertonic NaCl did not suppress 
renin secretion which remained elevated 
from the control at a rate of 365 ± 29 ng 
A-I/min (P < 0.05) (Fig. 2). This value was 
not significantly different from the level of 
463 ± 42 ng A-I/min obtained with the infu- 
sion of PGI2 alone (Fig. 2). During the re- 
covery period, renin secretion appeared to 
remain elevated but the change did not 
reach statistical significance at the 5% 



38 



PROSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



(N-6) «-P<OSfromC 


ttoCI 


ri 


■M»^ 


. • * . 1 


RENAL 




rh 




r 




rf 




T 




BIOOO ,30, 




















FLOW 


■ \ ^ 


















(ml/min) 
























1« 


MjCI 




« 

rh 


Hh 
















RENIN 












rh 


, i ,. 1 


SECRETION 




















(ng Al/min) 




















IW>- 


1 


















DOSE 0. 


I 


















POIa(lOr«g/)(B 


w<= 


1.0 




1.5 




J.5 




t.fl 



Fig. 2. Responses to intrarenal PGI2 and NaCl infu- 
sions in dogs with a single denervated nonfiltering kid- 
ney. Values are expressed as means ± SE. 

level. Renal blood flow significantly in- 
creased {P < 0.05) and RVR decreased {P 
< 0.05) during the administration of PGI2 
and these changes were unaffected by the 
administration of hypertonic NaCl (Fig. 2, 
Table II). As with the filtering kidney, RVP 
K concentration appeared to decrease 
slightly with the infusion of hypertonic 
NaCl (Table II), but the change was not 
significant. Finally, as in the first series, 
MAP remained unchanged throughout the 
experiment. 

Discussion. The results of the present 
study demonstrate the active participation 
of an intact renal tubular mechanism in the 



modulation of prostacyclin-stimulated renin 
release. In the present experiments and in 
agreement with previous reports from this 
laboratory (4), intrarenal illusion of non- 
hypotensive doses of PGI2 significantly 
stimulated renin release in dogs with either 
single filtering or nonfiltering kidneys. 
During this hyperreninemic state, the influ- 
ence of the renal tubular mechanism was 
evaluated by intrarenal infusion of hyper- 
tonic NaCl, an intervention that is well 
known to inhibit renin release (13, 16). Ac- 
cordingly, in our first series of experiments 
with filtering kidneys, the superimposition 
of hypertonic NaCl on the ongoing PGI2 
infusion significantly reduced renin secre- 
tion to the control level. This suppression 
was associated with an increment in RVP 
Na concentration, in the filtered load of 
sodium, and in sodium excretion. On the 
other hand, in dogs with a nonfiltering kid- 
ney, in an identical protocol NaCl infusion 
failed to attenuate significantly renin re- 
lease, even though the increment in RVP 
Na concentration was similar to the level 
achieved in dogs with filtering kidneys. 
Comparison of renin secretion during 
hypertonic NaCl infusion in the filtering 
and nonfiltering kidney series clearly indi- 
cates that in the absence of a tubular mech- 
anism, hypertonic NaCl was ineffective in 
reducing renin release. It seems reasonable 
to suggest that the renal tubular mechanism 
was mediated by the macula densa in the 
series with the filtering kidney. 
In the present study the nonfiltering kid- 



TABLE 11. Changes in Cardiovascular and Renal Hemodynamic Function during PGI2 and NaCl 
Infusion into Nonfiltering Kidneys of Anesthetized Dogs {N = 6) 



MAP mm Hg 
RVR mm Hg/ 

[ml/min] 
RVP Na meq/L 
RVP K meq/L 



C 

140 ±4 

1.90 ±0.28 
145 ± 1 
4.3 ±0.1 





PGI2 infusion at 1.0 or 1.5 x 10 " g/kg/min 




1.0 


NaCl 


1.5 


1.5 1 1.5 



1.02* ±0.15 
144 ± 1 
4.4 ±0.1 



0.91* ±0.12 
143 ± 1 
4.4 ±0.2 



0.91* ±0.13 
164*t ± 2 
4.1 ±0.2 



141 ±2 

0.86* ±0.14 
152* ± 1 
4.4 ± 0.2 



Note. Values are expressed as mean ± SE. Abbreviations are explained in the text. The asterisk (*) indicates 
statistically significant P value for the noted period compared with control. The dagger (t) indicates statistically 
significant P value for the noted period compared to the immediately preceding one. 



PROSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



39 



neys were denervated whereas the filtering 
kidneys were innervated. In the filtering 
kidneys, it seems unlikely that intrarenal 
NaCi infusion influenced renal nerve activ- 
ity with a subsequent effect on renin re- 
lease. In an earlier study by Shade et al. 
(13) in which the renal nerves were intact in 
both filtering and nonfiltering kidneys, a 
similar intrarenal infusion of hypertonic 
NaCl decreased renin secretion only in the 
filtering kidney. 

The renin secretion response during the 
recovery phase in the filtering and nonfil- 
tering kidney preparations was variable and 
requires additional comment. In both 
groups, renin secretion appeared to be ele- 
vated above control levels, but the incre- 
ments were not statistically significant at 
the 5% level for either group. Similarly, 
renin secretion in the recovery period was 
not different from PGI2 induced renin se- 
cretion prior to hypertonic NaCl infusion. It 
is possible that the observed increased 
variability in renin secretion during the re- 
covery phase was related to nonspecific 
secondary mechanisms, possibly volume 
expansion. Also, in the filtering kidney 
series RVP Na and sodium excretion re- 
mained elevated during the recovery period 
and this could have continued to attenuate 
partially the renin response to prostacyclin 
infusion. These findings were not unex- 
pected, since similar responses in the re- 
covery phase were observed by Shade et al, 
(13) using a similar protocol of hypertonic 
NaCl infusion in filtering and nonfiltering 
kidneys of anesthetized dogs with hyper- 
reninemia due to caval constriction. 

The bulk of the early evidence in the lit- 
erature (11, 12, 16) supports the view that 
an increased tubular sodium concentration 
or the load of sodium at the macula densa 
can attenuate renin secretion. More re- 
cently, studies by Kotchen et al, (17, 18) 
have postulated that chloride rather than 
sodium might be the important ion for this 
effect. In addition, there is the suggestion 
(11) that both sodium and chloride are in- 
volved in the macula densa sensor mecha- 
nism. Regardless of which ion(s) are de- 
tected by the macula densa, comparative 
studies with hypertonic NaCl infusion in 



the filtering and nonfiltering kidney (13) 
have provided evidence to support an im- 
portant role for the tubular system in the 
control of renin secretion. Inhibition of 
renin release by hypertonic NaCl has been 
demonstrated in normal anesthetized dogs 
with baseline renin secretion (16) as well as 
in dogs with hyperreninemia induced by 
caval constriction (13) or by norepinephrine 
(16). These observations and the present 
data are consistent with the idea that the 
macula densa can attenuate both basal and 
stimulated levels of renin secretion. The 
above considerations are particularly in- 
teresting when applied to prostaglandin- 
dependent renin release. Under appropriate 
conditions, it is possible that the natriuresis 
induced by various endogenous prostaglan- 
dins could reflect increased NaCl delivery 
to the macula densa and a feedback mecha- 
nism which functions as a brake on the in- 
crease in renin release. A similar postulate 
has been suggested in a recent review by 
Henrich (19). Studies by Higashihara ^/ al, 
(20) and Stokes (21) have suggested a 
PGE2-dependent inhibition of chloride 
reabsorption in the thick ascending limb of 
Henle, but the interpretation of the data has 
been questioned (22) and conclusive evi- 
dence of the action of prostaglandins on 
tubular sodium and chloride transport is 
lacking. 

The effects of prostacyclin on renal 
hemodynamic function and electrolyte 
excretion have been well documented and 
the present results are in agreement with 
previous studies (4, 10). In both the filtering 
and nonfiltering kidney series, the elevation 
in RBF produced by PGI2 infusion was not 
altered by the administration of hypertonic 
NaCl, while renin secretion was strikingly 
reduced in the filtering kidney preparation. 
This finding demonstrates that hypertonic 
NaCl did not inactivate or directly block the 
biological action of PGI2. Other inves- 
tigators have also reported lack of changes 
in renal hemodynamic function during in- 
trarenal NaCl infusion with similar pro- 
tocols in sodium replete dogs (16) or in dogs 
with caval constriction (13). 

In summary, hypertonic NaCl inhibited 
PGIo-stimulated renin secretion in the 



40 



PROSTAGLANDINS, THE MACULA DENSA, AND RENIN RELEASE 



single filtering kidney while hypertonic 
NaCl had no effect on PGl2-dependent 
renin release in the nonflltering kidney. 
Thus an intact tubular mechanism is essen- 
tial for this response. It is suggested that 
changes in tubular delivery of NaCl to the 
macula densa can modulate the hyp^r- 
reninemia induced by PGI2. 

We are grateful to Dr. John E. Pike of the Research 
Laboratories of the Upjohn Company for the supply of 
prostacyclin used in this study. We also wish to thank 
Mr. C. Gay, Mrs. G. Duflf. Ms. M. Flood, Ms. D. 
Pacropis, Mr. G. Garoutte, and Mr. D. Welch for their 
expert technical assistance. 

1. Bolgcr PM, Eisner GM, Romwell PW, Slotkoff 
LM. Effect of prostaglandin synthesis on renal 
function and renin in the dog. Nature (London) 
259:244-245, 1976. 

2. Gerber JG, Branch RA, Nies AS, Gerkens JF, 
Shand DG, HoUifield J, Gates J A. Prostaglandins 
and renin release. II. Assessment of renin secre- 
tion following infusion of PGI2, Ez and D^ into the 
renal artery of anesthetized dogs. Prostaglandins 
15:81-88, 1978. 

3. Seymour AA, 2^hr JE. Influence of renal prosta- 
glandin synthesis on renin control mechanisms in 
the dog. Circ Res 45:13-25, 1979. 

4. Seymour A A. Davis JO, Freeman RH, DeForrest 
JM, Rowe HP, Williams GM. Renin release from 
filtering and nonflltering kidneys stimulated by 
PGI2 and PGD2. Amer J Physiol 237(4):F285- 
F290, 1979. 

5. Yun J, KeUy G, Bartter FC, Smith H, Jr. Role of 
prostaglandins in the control of renin secretion in 
the dog. Circ Res 40:459-464, 1977. 

6. Gates J A, Whorton AR, Gerkens JG, Branch RA, 
Hollifield JW, FrOlich JC. The participation of 
prostaglandins in the control of renin release. Fed 
Proc 38:72-74, 1979. 

7. Frolich JC. Prostaglandins: Role in renin regula- 
tion and mediation of antihypertensive drug ef- 
fects. Arch Int Pharmacodyn Ther Suppl:213- 
224, 1980. 

8. Suzuki S, Franco Saenz R, Tan SY, Mulrow PW. 
Direct action of prostaglandins on renin release 
from rat renal cortical slices. Proc Soc Exp Biol 
Med 166:484-488, 1981. 

9. Whorton AR, Misorpo K, Hollifield J, Frolich JC, 
Inagami T, Gates J A. Prostaglandins and renin re- 
lease. I. Stimulation of renin release from rabbit 



renal cortical slices by PGIz. Prostaglandins 
14:1095-1104, 1977. 

10. Bolger PM. Eisner GM, Romwell PW, Slotkofif 
LM. Renal actions of prostacyclin. Nature (Lon- 
don) 271:467-469, 1978. 

11. Davis JO, Freeman RH. Mechanisms regulating 
renin release. Physiol Rev 56:1-56, 1976. 

12. Freeman RH, Davis JO, Gotshall RW, Johnson 
JA, Spielman WS. The signal perceived by the 
macula densa during changes in renin release. 
Circ Res 35:307-315, 1974. 

13. Shade RE, Davis JO, Johnson J A, Witty RT. Ef- 
fects of renal arterial infusion of sodium and 
potassium on renin secretion in the dog. Circ 
Res 31:719-727, 1972. 

14. Blaine EG, Davis JO, Witty RT. Renin release 
after hemorrhage and after suprarenal aortic con- 
striction in dogs without sodium delivery to the 
macula densa. Circ Res 28:1081-1089, 1979. 

15. Sealey JE, Laragh JH, Gerten-Bames J, Accto 
RM. The measurement of plasma renin activity in 
man. In: Hypertension Manual. Yorke Medical 
Books/Dunn- Donnelley, New York, pp621- 
640, 1974. 

16. Nash FD, Rostorfer HH, Bailie MD, Wathen RL, 
Schneider EG. Renin release in relation to renal 
sodium load and dissociation from hemodynamic 
changes. Circ Res 22:473-487, 1968. 

17. Kotchen TA, Galla JH, Luke RG. Contribution of 
chloride to the inhibition of plasma renin by 
sodium chloride in the rat. Kidney Int 13:201- 
207, 1978. 

18. Kotchen TA, Krzyzaniak KE, Anderson JE, 
Ernst CB, Galla JH, Luke RG. Inhibition of renin 
secretion by HCI is related to chloride in both dog 
and rat. Amer J Physiol 239(8): F44-F49, 1980. 

19. Henrich WL. Role of prostaglandins in renin se- 
cretion. Kidney Int 19:822-830, 1981. 

20. Higashihara E, Stokes JG, Kokko JP, Campbell 
WB, DuBose TD, Jr. Cortical and papillary mi- 
cropuncture examination of chloride transport in 
segments of the rat kidney during inhibition of 
prostaglandin production. J Clin Invest 64:1277- 
1287, 1979. 

21. Stokes JB. Effect of prostaglandin E2 on chloride 
transport across the rabbit thick ascending limb of 
Henle. J Clin Invest 64:495-502, 1979. 

22. Fine LG, Kirschenbaum MA. Absence of direct 
effects of prostaglandins on sodium chloride 
transport in the mammalian nephron. Kidney Int 
19:797-801, 1981. 

Received January 4, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 41-45 (1982) 



Differences in the Reported Frequencies of Cleft Lip Plus Cleft Lip and Palate in 
Asians Born in Hawaii and the Continental United States^ (41474) 

MARVIN L. TYAISP 

Medical and Research Services, Wadsworth VA Medical Center, Wilshire and Sawtelle, Los Angeles, 
California 90073, and The Department of Medicine, School of Medicine. University of California, 

Los Angeles, California 90024 



Abstract. The etiology of cleft Up with or without cleft palate (CL/CP) is not known but 
present evidence suggests that both genetic and environnoental foctors act to determine 
susceptibility. Race has been reported to exert a strong influence on the incidence of CL/CP; 
e.g., the frequency of CL/CP is neaiiy twice as high among Japanese bom in Japan or Hawaii 
as it is among Caucasians in Hawaii, Western Europe, and North America, and the risk in 
American blacks is one-half that in Caucasians. This notwithstanding, a survey of Los 
Angeles Hospitals and Clinics for families with fsicial clefting revealed no Orientals despite 
the fact that this group constitutes 6% of the population. To investigate the possibility that 
the rate of clefting had declined in Orientals* data was obtained from birth certificates from 
Hawaii, California, and New York, and from the USPHS Birth Oefects Monitoring Program. 
The results suggest that the frequency of CL/CP but not isolated CP may be significantly 
lower among Japanese and other Orientals bom in Califomia and New York than among 
those bom in Japan or Hawaii. This implies that environmental foe tors play a m^jor role in 
determining the frequency of CL/CP in this racial group in the Orient and Hawaii. 



Isolated cleft palate (CP) and cleft lip 
with or without cleft palate (CL/CP) are 
congenital malformations of unknown 
etiology. Both anomalies tend to recur in 
families and also to vary in incidence with 
season and year of conception, and parental 
social class, age, parity, ethnic origin, and 
place of residence during pregnancy (1-4). 
It is not clear whether the variations in the 
frequency of CP and CL/CP observed 
among populations and races are predomi- 
nantly genetic or environmental in origin, 
because in past studies most of the groups 
compared differed in both respects. 

The influence of race on the incidence of 
CP and CL/CP appears to be especially 
strong (5). The frequency of cleft lip with or 
without cleft palate has been reported to be 
nearly twice as high among Japanese bom 
in Japan or Hawaii as it is among Cauca- 
sians bom in Hawaii, Europe, or North 



» Supported in part by USPHS Grant DE 05165 and 
VA Medical Research Funds. 
' To whom all correspondence should be addressed. 



America (6, 7) and the frequency of CP in 
blacks is approximately one-half that in 
American Caucasians (1). This notwith- 
standing, a survey of Los Angeles Hos- 
pitals and Clinics for families with fa- 
cial clefting revealed no Orientals even 
though this group constitutes 6% of the 
population. It was concluded that either 
Oriental patients were going to private 
physicians not affiliated with mainstream 
medical institutions or the incidence of 
clefting was low. 

In an effort to determine if the frequency 
of facial clefting in the Oriental population 
in Califomia had truly declined, data from 
birth certificates were obtained from the 
States of Hawaii, California, and New 
York, and hospital discharge diagnoses 
were secured from the USPHS Birth De- 
fects Monitoring Program. The results 
suggest that the frequency of CL/CP but not 
CP may be significantly lower among 
Japanese and other Orientals born in 
Califomia and New York than among those 
bom in Japan or Hawaii. This implies that 
environmental factors play a major role in 



41 

0037-9727/82/090041^5$01 .00/0 
Copyrifht © 1982 b^ \!kMt V»cmVi Vk ^xv^eoM 



42 



CLEFT LIP AND PALATE 



determining the frequency of CL/CP in this 
racial group in the Orient and Hawaii. 

Methods. Computer printouts giving total 
live births, and the number of isolated cleft 
palates, cleft lips, and cleft lips plus cleft 
palates with the race of the mother and 
father were obtained from the Departments 
of Health of the States of Hawaii, Califor- 
nia, and New York. In addition, data were 
obtained from the Birth Defects Monitoring 
Program, USPHS Center for Disease Con- 
trol, Atlanta, Georgia. The data from the 
USPHS CDC do not contain information on 
the race of mother or father, and all Orien- 
tals are recorded as "Asians." Approxi- 
mately 20% of all * 'Asian" births were re- 
ported from the State of Hawaii. 

Isolated cleft palate and cleft lip with or 
without cleft palate are considered to have 
distinct genetic and environmental etiol- 
ogies (1, 7). Because ascertainment varies 
between studies, comparisons of frequen- 
cies of CP and CL/CP were made between 
races within each study only. It is assumed 
that nonascertainment within a study will 
be equally distributed over racial subgroups 
of the population as was the case in Hawaii 
as reported by Ching and Chung (7). With 
the exception of the data from the Birth 
Defects Monitoring Program the informa- 
tion reported is derived from incrosses 
only. 

Statistical analyses were done by Leslie 
Bernstein, Ph.D. The Mantel -Haenszel 
test (8) was used to test the hypothesis of 
equal incidence (risk ratio = 1) among 
Caucasians and Japanese or Asians. 

Results. Isolated cleft palate among 
Caucasians and Japanese born in Hawaii 
or California {Table /). The frequencies of 
isolated cleft palate among Japanese in 
Hawaii and California were consistently 
lower than those found among Caucasians; 
however, the differences were not statisti- 
cally significant. 

Cleft lip and cleft lip plus cleft palate 
among Caucasians and Japanese born in 
Hawaii or California {Table I). The relative 
risk of Japanese in Hawaii producing prog- 
eny with CL/CP was 1.85 to 2.41 times 
higher than that in Caucasians; however, in 



u 

:3 

h 

ti 

D = 

J < 

t ^ 

ts 

< c/) 

< - 

< < 
ft- ^ 

u. < 

S o 

CQ 
< 



.2 -r 
"5 c 



u2 



u2 



4> 

> 



0.0 
US 



4 •« 



2| 



u 






O - 

z.- 



ifl 



^ — O O 



d 



28 SS8 

do ^ d 



00 o 



00 m 



00 — 

K — ^ 

O 00 

00 o 



5 « w ;, 

II 'c I I 

•= 00 '^ t 00 Tf 

^ o o ^t2 (TV (TV 

rt -3 

3= U 



xu 



SB 

BB 

CO CO 



XX 
o o 
c c 
E £ 

II 

Q, Q, 



Q, Q, 

a a 

3 3 






t: c 

lo IE 

t^ iS iS v 
a> cd eo 



CLEFT LIP AND PALATE 



43 



California during roughly the same period 
the risk was essentially the same. 

Isolated cleft palate among Caucasians 
and Asians born in Hawaii or in the United 
States {Table II), The relative risk of 
Chinese, Japanese, or Filipinos, either as 
individual races where those data were 
available or as ''Asians,'' producing prog- 
eny with CP in Hawaii, California, New 
York or in other areas of the United States 
was consistently but not significantly below 
that of Caucasians. 

Cleft lip and cleft lip plus cleft palate 
among Caucasians and Asians born in 
Hawaii or in the United States (Table II). 
The relative risk of CL/CP was higher in all 
Asian racial groups in Hawaii (1.S3 to 2.41, 
individual data not shown), but in Califor- 
nia and New York the frequencies were 
somewhat lower than that of Caucasians. 
The relative risk of CL/CP in Asians (1.20) 
from the USPHS data (1970- 1979) was not 
significantly higher than that of Caucasians 
even though approximately 20% of the 
Asian births were reported from Hawaii. 
(From this one would reason that the risk 
among Asians bom in the continental USA 
is lower than 1.20.) 

Discussion. Opinions differ on whether 
the variations in the frequencies of facial 
clefting between racial groups are due 
primarily to differences in the environment 
(9) or in the population frequency of partic- 
ular combinations of genes which influence 
development (5). The findings presented 
here support the view that the variations in 
frequencies of CL/CP noted between 
Caucasians and Japanese and other Orien- 
tals born in Hawaii and the United States 
are secondary to environmental changes. 

The few studies which have reported 
differences in the frequency of clefts among 
races agree that clefts are seen much less 
frequently among blacks than among 
Caucasians from the same geographic area. 
It has been reported that only seven cleft 
cases were observed among 12,520 black 
births (0.55/1000) at Johns Hopkins Hospi- 
tal. During the same period there were 17 
cases among 15,656 Caucasian births 
(1.06/1000) (10). Other studies based on 



a 
z 

o 

< 

< 

*** 111 

U a: 

Q O 

z = 

< < 

a. ^ 



U. z 
cu - 

UJ ^ 

U < 

<i 



CQ 
< 



u 






u2 



u2 



I. 

CO ^ 



ft- 2 



i S? 



ft* 2 
U 






U 



1 1 



— 8 r^ 
fn ^ ri 



d 



d 



^ o -^ 



d 



^ d 



d d 



d d 



r^ 


rr% 


•^ 


«M 


vO 


ro 


«o 


n 


d 


d 


d 


d 


•^ 


S 


% 


1 


ON 




















-^ 


m 


•^ 


i 




s 



^. - On On ^ O^ 



•3 "^ 5 -^ IT- o p** i^. 
eo r> ,o r». au r». r- 

^ONJtJOvCuov i:> 

CO — -3 — CO ~ ^. 



ih 

2 



-if 
z 

o 



e * 

2 c 

•5 I 

o « 
• <*5 IS. 

^ o 

« 6 E 
^ o o 
^(h (£: 

•^Q Q 



44 



CLEFT LIP AND PALATE 



birth certificate data have reported consis- 
tently lower attack rates of facial clefting 
among American blacks than among Cau- 
casians (11-13) and this relationship is 
maintained despite geographic and cultural 
change (2). Although these studies strongly 
suggest that genetic differences primarily 
account for the variations in attack rates for 
CP and CL/CP between Blacks and Cauca- 
sians, as with all studies of this type the 
differences may be caused by unrecognized 
variations in ascertainment, local environ- 
ment, racial classification, or diagnostic 
criteria. Some of the problems associated 
with the use of buth certificates in the epi- 
demiologic assessment of facial clefts are 
addressed in Refs. 16 and 17. 

Ching and Chung (7) reported that the 
Caucasian incidence of oral clefts in Hawaii 
was approximately the median value found 
among other Caucasoid populations studied 
(14) and that the incidence of CL/CP for 
Japanese was significantly higher and in- 
termediate between the incidences esti- 
mated by Neel (5), Koguchi (6), and 
Kobayashi (15) for Japanese in Japan sug- 
gesting that the racial frequencies were not 
altered significantly by the migration of the 
Japanese and Caucasians to Hawaii (the 
frequency of CP among Japanese bom in 
Hawaii was slightly higher than that esti- 
mated for Japanese in Japan). The data 
presented here support those findings but 
suggest strongly that when the Japanese 
and perhaps other Asians migrate to 
California, New York, or other areas of 
mainland United States the frequency of 
CL/CP but not CP decreases to approxi- 
mate that noted in Caucasians. 

There are three explanations for the dif- 
ferences observed in the frequency of 
CL/CP among the Japanese living in Hawaii 
or California and among other Asians living 
in Hawaii and the United States. There may 
be cultural differences in the reporting fre- 
quency (16, 17), migration to the mainland 
may not be representative of the entire 
population or there may have been a true 
decrease in the occurrence of this anomaly 
in these racial groups as a function of time 
or assimilation into a different culture and 
environment. At present there is no evi- 



dence in support of the first two pos- 
sibilities but observations exist which sup- 
port the concept that attack rates for facial 
clefting in certain racial groups do vary with 
geography and environment (1, 2, 4). 
Prominent among changes which occur 
with migration is diet. Oriental and Ha- 
waiian diets tend to be lower in fat and 
animal protein and higher in fish, vegeta- 
bles and vitamin A than does the average 
western diet (18, 19). This may be of spe- 
cific interest because it is known that high 
dietary levels of vitamin A increase sus- 
ceptibility to cleft palate in mice bearing 
H-2D^ alleles in their major histocompati- 
bility complex (20). 

Finally, the observation that the fre- 
quency of isolated cleft palate among the 
Japanese did not vary with time and place 
as did that of CL/CP tends to support the 
view that these anomalies have distinct 
etiologies. 

I wish to thank Joan P. Cooney, Office of Biostatis- 
tics. New York State Department of Health, Frank D. 
Norris, Center for Health Statistics, California De- 
partment of Health, Thomas A. Burch, M.D., M.P.H., 
Chief, Research and Statistics Office, Hawaii Depart- 
ment of Health, and Larry Edmonds, Center for Dis- 
ease Control, Atlanta, Georgia, for their efforts in pro- 
viding the data used in these studies. 



1. Greene JC. Epidemiology of congenital clefts of 
the lip and palate. Public Health Reports, Public 
Health Service, U.S. Department of Health, Edu- 
cation and Welfare, Vol 78:pp589-602. 1963. 

2. Leek I. Ethnic differences in the incidence of 
malformations following nugration. Brit J Prev 
Soc Med 23:166-173, 1969. 

3. Wehrung DA, Hay S. A study of seasonal inci- 
dence of congenital malformations in the United 
States. Brit J Prev Soc Med 24:24-32, 1970. 

4. Saxen I. Epidemiology of cleft lip and palate. Brit 
J Prev Soc Med 29:103-110, 1975. 

5. Neel JV. A study of major congenital defects in 
Japanese infants. AmerJ Hum Genet 10:398-445, 
1958. 

6. Koguchi H. Population data on cleft lip and cleft 
palate in the Japanese. In Melnick M, Bixler D, 
Shields ED, eds. Etiology of Cleft Lip and Clefl 
Palate. New York, Liss, pp297-323, 1980. 

7. Ching GSH, Chung CS. A genetic study of cleft lip 
and palate in Hawaii. Amer J Hum Genet 
26:162-176, 1974. 



CLEFT LIP AND PALATE 



45 



8. Mantel N, Haenszel W. Statistical aspects of the 
analysis of data from retrospective studies of dis- 
ease. J Nat Cancer Inst 22:719-748, 1959. 

9. Morton NE. Genetics of inter-racial crosses in 
Hawaii. Eugen Q 9:23-29, 1962. 

10. Davis JS. The incidence of congenital clefts of the 
lip and palate. Ann Surg 80:363-374, 1924. 

11. Loretz W, Westmoreland WW, Richards LF. A 
study of cleft lip and cleft palate births in Califor- 
nia, 1955. Amer J Public Health 51:873-877, 
1961. 

12. Grace LG. Frequency of occurrence of cleft pal- 
ates and harelips. J Dent Res 22:495-497, 1943. 

13. Lutz KR, Moor FB. A study of factors in the oc- 
currence of cleft palate. J Speech Hearing Dis 
20:271-276, 1955. 

14. Leek I. The etiology of human malformations: In- 
sights from epidemiology. Teratology 5:303-314, 
1972. 

15. Kobayoshi Y. A genetic study on harelip and cleft 
palate. Jap J Hum Genet 3:73-107, 1958. 

16. Mackeprang M, Hay S, Lunde AS. Completeness 



and accuracy of reporting of malformations on 
birth certificates. Health Services and Mental 
Health Administration Health Reports 87:43-49, 
1972. 

17. Meskin LH, Pruzansky S. Validity of the birth 
certificate in the epidemiologic assessment of fe- 
cial clefts. J Dent Res 46:1456-1459, 1967. 

18. Hankin JH, Nomura A, Rhoads GG. Dietary pat- 
terns among men of Japanese ancestry in Hawaii. 
Cancer Res 35:3259-3264, 1975. 

19. Kagan A, et al. Epidemiologic studies of coronary 
heart disease and stroke in Japanese men living in 
Japan, Hawaii and California; demographic, 
physical, dietary and biochemical characteristics. 
J Chron Dis 27:345-364, 1974. 

20. Tyan ML, Miller KK. Genetic and environmental 
factors in cortisone-induced cleft palate. Proc Soc 
Exp Biol Med 158:618-621, 1978. 



Received January 14, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 46-49 (1982) 



Ovine Maternal and Fetal Circulatory Responses to an Endoperoxide 

Analog (41475) 

DAVID B. SCHWARTZ,' TERRANCE M. PHERNETTON, MICHAEL K. STOCK, 

AND JOHN H. G. RANKIN 

Departments of OhstetrUs-Gynecology and Physiology, University of Wisconsin, and Wisconsin Perinatal 
Center, Madison General Hospital, Madison, Wisconsin 53715 



Abstract. There are extensive data on the circulatory responses to prostaglandins during 
pregnancy, but little is known about the precursor endoperoxides. The endoperoxide analog 
(15S)-hydroxy-9a,l la-(epoxymethano)prosta-5Z,13E-dienoic acid (EPAi) was used to eval- 
uate the circulatory effects of endoperoxides in pregnant sheep. Ten near-term ewes and 
their fetuses were chronically catheterized to permit the measurement of regional blood 
flows by the radioactive microsphere method. In Ave sheep a fetal IV bolus injection of 12.5 
/Ag/kg EPA, produced a significant increase in fetal blood pressure from 43 to 51 mm Hg, 
vascular resistance of the cotyledons from 0.06 to 0.09 PRU kg', membranes from 0.84 to 
3.15 PRUkg ', and kidneys from 1.18 to 1.94 PRUkg '. In Ave sheep a 10-min maternal 
infusion of 0.5 /xg/kg/min EPA, produced a significant increase in blood pressure from 89.0 
to 115.0 mm Hg. resistance of the uterus from 0.49 to 1.10 PRU. and kidneys from 0.13 to 
0.20 PRU. There was no significant change in maternal cotyledonary resistance. It is evident 
from these results that EPA, causes widespread vasoconstriction. In the placenta, however, 
there is vasoconstriction on the fetal side but the resistance of the maternal vascular bed is 
unaltered. 



The endoperoxides are intermediaries in 
the formation of the prostaglandins, are 
biologically active, and exert strong and in- 
dependent pharmacologic and physiologic 
effects on vascular and respiratory smooth 
muscle (1). For several years we have in- 
vestigated the role of prostaglandins in reg- 
ulating and modulating placental blood 
flows (2-6). The actions of the endoper- 
oxides are not well documented. The half- 
life of the endoperoxides is 4 to 5 min (7, 
8) and some of the circulatory effects at- 
tributed to the prostaglandins may well be 
due to endoperoxide action (1). The en- 
doperoxides derived from arachidonic acid 
are prostaglandins G2 (PGG2) and H2 
(PGH2). These are powerful vasoconstric- 
tors, bronchoconstrictors, and platelet ag- 
gregating agents (7). As there is relatively 
little information on maternal and fetal 
cotyledonary responses to endoperoxides, 
this study was designed to evaluate these 
responses in pregnant sheep. The substance 



' To whom correspondence should be addressed. 



used in the experiments was the stable en- 
doperoxide analog (15S)-hydroxy-9a,lla- 
(epoxymethano)prosta-5Z, 1 3E-dienoic acid 
(EPA,) (9). 

Materials and Methods. Ten cross-bred 
sheep were used in this study. Surgical 
preparation was performed between Day 
125 and Day 128 of gestation. Anesthesia 
was induced with an intramuscular injec- 
tion of ketamine (10 mg/kg) and atropine 
(0.6 mg). Anesthesia was maintained with 
a continuous intravenous infusion of ket- 
amine in normal saline (10 mg/min). A ma- 
ternal left ventricular catheter was placed 
via the right carotid artery with the use 
of local xylocaine to supplement the par- 
enteral anesthesia. A polyethylene cath- 
eter (PE200), with an inner polyvinyl cath- 
eter (i.d. 0.5 mm), was inserted into the 
common carotid artery and advanced into 
the left ventricle. The catheter position was 
confirmed by the characteristic pressure 
pattern. Polyvinyl catheters (i.d. 0.5 mm) 
were placed in superficial peripheral ar- 
teries of both maternal hindlimbs and ad- 
vanced for 20 cm into the femoral artery. 



46 
}2/090046-04$01.0()/0 
982 by Ihc StH:icty for Kxpcnmental Biology and Medicine. 



PREGNANT OVINE VASCULAR RESPONSE TO EPA, 



47 



The maternal abdomen was opened in the 
midline, the uterus palpated, and a fetal 
hindlimb identified. The fetal hindlimb ar- 
tery and vein were catheterized as has been 
previously described (10). The uterus and 
abdomen were closed in layers and the 
catheters were led via a subcutaneous tun- 
nel to a flank incision where they were 
maintained in a pouch. The left ventricular 
catheter was secured under an elastic ban- 
dage placed around the neck. 

The experiments were performed 2 days 
after the surgical procedures to allow for 
recovery from surgical stress. The ewe was 
placed in a stanchion in the laboratory and 
permitted to stabilize for 2 hr. The maternal 
and fetal mean arterial blood pressures 
were measured with a Statham P23Db 
transducer placed at the level of the 
scapulohumeral joint. These parameters 
were displayed on a Beckman R411 record- 
er with an EO-18 oscilloscope. Maternal 
and fetal arterial blood gas analysis was 
performed prior to and at the conclusion of 
each experiment. 

The EPA, was prepared as a 1 mg/ml 
stock solution in ethanol and stored at 4^. 
This solution is stable for 6 months at 4° (9). 
On the days of the experiments dilutions 
were made from the stock solution and kept 
on ice until the time of injection. 

In five of the experiments, EPAi was in- 
fused into the mother and the maternal cir- 
culatory responses were measured by the 
radioactive microsphere method. In the 
other five experiments a bolus injection of 
the same substance was administered to the 
fetus and regional blood flows were mea- 
sured with the same techniques. 

In the maternal series, an infusion of O.S 
/ng/kg/min was administered via the left 
ventricular catheter over a 10-min period. 
Blood flows to individual organs of the 
mother were measured by injecting radio- 
active microspheres into the left ventricle 
while withdrawing an integrated arterial 
sample from the femoral arterial cathe- 
ter at a rate of 2.06 ml/min for 1.5 min in a 
manner previously described (11). A con- 
trol observation was made of organ blood 
flows with the initial isotope injection prior 
to starting the inftision, and a second ob- 
servation was made just before the conclu- 



sion of the 10-min infusion. Microspheres 
(3M Company and NEN) were 15 fxm in 
diameter and were labeled with one of the 
following isotopes: ^«Sc, *^Co, »*Sr, »^Cd. 
Each injection contained approximately 
1.25 million spheres. 

In the fetal series of experiments a bolus 
of 12.5 Mg/kg EPA, was injected into the 
fetal hindlimb vein. Regional blood flows 
were measured in the control condition (C) 
and at 5 min after the administration of 
EPA, (T). 

At the conclusion of the experiment, the 
ewe was sacrificed and the kidneys, uterus, 
and contents were removed. The cotyle- 
dons were dissected free from the uterus, 
and cotyledonary and noncotyledonary tis- 
sues were separately assayed. The fetal 
kidneys were also removed. All tissues 
were weighed and homogenized with a 
measured volume of water in a Waring 
blender. Five aliquots were taken of each 
homogenate. Each aliquot had a volume of 
approximately 2 ml and was placed in a 
preweighed, wide-mouth counting vial 
which was weighed again. The homogenate 
reached a height of approximately 1 cm in 
the vial. Details of these techniques are 
available in an earlier publication (11). A 
sample progression was established for 
each experiment in which standard vials, 
appropriate to each isotope used in the ex- 
periment, were followed by vials containing 
the blood samples obtained from the inte- 
grated arterial sample withdrawal. These 
were followed by the vials containing the 
homogenized tissue. All measurements of 
radioactivity were made with a three- 
channel, well-type automatic gamma count- 
er (Model 1185 Nuclear Chicago). In ex- 
periments in which four isotopes were used, 
the windows of the gamma counting system 
were reset and a second pass was made to 
detect the fourth isotope. The output of 
the gamma counter was on paper tape 
which was fed into a Univac 1110 computer 
via an interactive terminal in our laborato- 
ries. Data reduction was performed solving 
the appropriate algorithms by iterative 
matrix procedures and programs supplied 
by our laboratories. The spillover of each 
isotope in the other channels was deter- 
mined from the information obtained from 



48 



PREGNANT OVINE VASCULAR RESPONSE TO EPA, 



the standard vials, as was the number of 
counts per minute per microsphere at that 
time. The data were reduced to the number 
of spheres per vial of homogenized tissue. 
The number of microspheres per gram of 
tissue homogenate was considered to be the 
average of the number of microspheres per 
gram observed in the five vials. Blood flow 
was determined with the equations given 
by Makowski et al. (12). The resistance of 
the maternal organs was defined as the 
mean maternal arterial blood pressure di- 
vided by the organ blood flow. The resist- 
ance of subdiaphragmatic fetal organs was 
defined as mean fetal arterial pressure 
minus mean fetal venous pressure, divided 
by the organ blood flow. Comparisons be- 
tween the control and test observations 
were made using a paired t test. Results 
were expressed as mean ± SEM. 

Results. Blood gas analysis indicated a 
mean maternal arterial pH of 7.50 ± 0.12, 
PCOt of 25.8 ± 3 mm Hg and PO2 of 75.9 ± 
8 mm Hg. The blood drawn from the fetal 
hindlimb artery had a mean pH of 7.43 ± 
0.06, POa of 19.8 ± 2 mm Hg, and PCO2 of 
34.6 ± 5 mm Hg. 

Maternal scries. The dose of EPA| cho- 
sen, 0.5 /Ltg/kg/min, was similar to the dos- 
age regime used by WaSvSerman (8) in 
studying bronchoconstrictor effects in 
dogs. Observations were obtained in four 
single and one twin pregnancy. The mean 
fetal weight was 3.4 ± 0.2 kg. The re- 
sponses to EPA, were as follows. The ma- 
ternal bliH^d pressure changed from 88 ± 5 
to 1 15 t 5 mm Hg (P ^ 0.002). This change 
was apparent within 20-40 sec of com- 
mencing the infusion and a hypertensive 
effect was noted to last for 10-15 min after 
completion of the infusion. The cotyledon- 
ar\ bkH>d flow was 165 ± 34 (mL'min)/kg 
fetus hcfone the infusion and 168 ± 24 (ml 
min V kg fetus during the infusion (NS). The 
renal bUxHl tlow was 688 ± 24 ml min be- 
fore the infusion and 621 it 62 ml min dur- 
ing the infusion iNS). During the infusion of 
FPA, ihc noncot\lcdonar\ blood tlow 
changed fix>m ;i; - 42 to \20 t. 30 ml min 
{P -. 0.05>. When assistance was calculated 
i( was found that ihc cot\ledonary asis- 
tancc did noi change during the infusion of 
l-PA^. The noncotyledonary uterine resis- 



tance changed from 0.49 ± 0.08 to 1.19 ± 
0.22 PRU (P < 0.01), and renal resistance 
changed from 0.13 ± 0.01 to 0.20 ± 0.03 
PRU {P < 0.025). These data indicate that 
EPA, causes maternal hypertension and 
vasoconstriction in the renal and non- 
cotyledonary circulations. 

Fetal series. The fetal dose of EPA, was 
chosen empirically to produce a consistent 
and obvious effect on the fetal blood pres- 
sure. Observations were obtained on five 
fetuses in five sheep. The mean fetal weight 
was 3.9 ± 0.3 kg. The injection of EPAi 
changed the mean arterial blood pressure 
from 43 ± 1 to 51 ± 1 mm Hg (P < 0.01). 
The cotyledonary blood flow was 180 ± 22 
(ml/min)/kg in the control condition and 143 
± 14 (ml/min)/kg after EPA, (NS). EPAi 
caused the membrane blood flow to change 
from 14 ± 1 to 5 ± 1 (ml/min)/kg (P < 0.01) 
and the renal blood flow to change from 10 
± 0.6 to 7 ± 0.8 (ml/min)/kg {P < 0.02). 
From these data the following resistances 
were calculated. The cotyledonary resis- 
tance changed from 0.06 ± 0.01 to 0.09 ± 
0.01 PRU/kg fetus (P < 0.01), the mem- 
brane resistance changed from 0.84 ± 0.14 
to 3. 15 ± 0.70 PRU/kg fetus {P < 0.01), and 
the renal resistance changed from 1.18 ± 
0. 10 to 1 .94 ± 0.32 PRU/kg fetus (P < 0.02). 
EPA, therefore resulted in fetal hyperten- 
sion and vasoconstriction in the cotyledon- 
ary, membranous, and renal vascular beds. 

Discussion. The radioactive microsphere 
method for measuring organ blood flow is a 
technique which has been well established 
(12-14). Sources of error are described by 
Buckberg et al. (13). The primary error is 
that of having insufficient microspheres in 
the assay sample. This problem was avoid- 
ed by counting several samples of the in- 
dividual tissues, and all the arterial blood 
samples and tissue samples contained more 
than 400 microspheres. 

The cyclic endoperoxides PGGj and 
PGHt are the initial intermediary metabo- 
lites produced by the cyclooxygenase 
catabolism of arachidonic acid. These arc 
the precursor substrates for the production 
of PGU, PGEj, PGF^i, PGDj, and the 
thri>mK>\anes. The biological half-life of 
the endoperoxides is 4-5 min and they 
ha\e been demonstrated to have powerful 



PREGNANT OVINE VASCULAR RESPONSE TO EPA, 



49 



constrictor effects on the vascular and air- 
way smooth muscle which is independent 
of the stable prostaglandins (8, IS, 16). The 
methylene derivatives of PGH2 (cyclic 
ethers) are called cyclic endoperoxide 
analogs 1 and 2 (8, 17). These substances 
are stable in aqueous solution and have 
been shown to be more potent smooth mus- 
cle stimulants than the prostaglandins (8). 
In both series of experiments the pro- 
found vasoconstrictor effects of EPAi were 
clearly evident. These effects were noted to 
be consistent in all tissues tested except for 
the maternal cotyledonary circulation, in 
which the blood flow and resistance were 
unaltered. The fetal placental circulation 
was sensitive to the vasoconstrictor effects 
of EPAi and demonstrated a significant in- 
crease in vascular resistance. It may be ar- 
gued that these differing responses of the 
two placental circulations were a function 
of the dosage and method of administering 
EPAi in the two sets of experiments. As all 
the other organs that were studied demon- 
strated significant vasoconstriction, the 
failure of the maternal cotyledonary circu- 
lation to respond in a similar fashion would 
indicate an absolute or relative insensitivity 
of this particular circulatory bed to the ef- 
fects of EPAi. The opposing effects on the 
two placental circulations is not a unique 
response. It has been observed in similar 
experiments using prostaglandin I2 (6). PGI2 
was noted to cause significant changes in 
maternal cotyledonary blood flow while fetal 
cotyledonary blood flow was unaltered. As 
has been suggested for prostaglandin Ij, the 
anomalous responses of the placental vas- 
culature to EPA, may be the result of an in- 
teraction with other vasoactive substances 
such as circulating catecholamines. 

We wish to thank Dr. John Pike of the Upjohn Com- 
pany for his cooperation in providing the EPAi for the 
conduct of these experiments and Mrs. Ruth Ledin for 
her help with the preparation of the manuscript. 

1. Samuelsson B. Prostaglandin endoperoxides and 
thromboxanes: Role in platelets and in vascular 
and respiratory smooth muscle. Acta Biol Med 
Germ 35:1055-1063, 1976. 

2. Rankin JHG, Phemetton TM. Effect of prosta- 
glandin Ez on ovine maternal placental blood flow. 
Amer J Physiol 231:754-759, 1976. 

3. Rankin JHG, Phemetton TM. Circulatory re- 



sponses of the near-term sheep fetus to prosta- 
glandin E2. Amer J Physiol 231:760-765, 1976. 

4. Rankin JHG. Role of prostaglandins in the main- 
tenance of the placental circulation. In: Coceani 
F, Olley PM, eds. Advances in Prostaglandin and 
Thromboxane Research. New York, Raven Press, 
Vol 4:pp 261-269, 1978. 

5. Rankin JHG, McLaughlin MK. The regulation of 
the placental blood flows. J Dev Physiol 1:3-30, 
1979. 

6. Rankin JHG, Phemetton TM, Anderson DF, 
Berssenbrugge AD. Efifect of prostaglandin U on 
ovine placental vasculature. J Dev Physiol 
1:151-160, 1979. 

7. Hillier K. Prostaglandins and thromboxanes: 
Pharmacologic and biosynthetic aspects. Semin 
Perinatol 2:197-210, 1978. 

8. Wasserman MA. Bronchopulmonary pharmacol- 
ogy of some prostaglandin endoperoxides and 
analogs in the dog. Eur J Pharmacol 36:103-114, 
1976. 

9. Bundy GL. The synthesis of prostaglandin en- 
doperoxide analogs. Tetrahedron Lett 24:1957- 
1960, 1975. 

10. Rankin JHG, Gresham EL, Battaglia FC, 
Makowski EL, Meschia G. Measurement of fetal 
renal inulin clearance in a chronic sheep prepara- 
tion. J Appl Physiol 32:129-133, 1972. 

11. Buss DD, Bisgard GE, Rawlings CA, Rankin 
JHG. Uteroplacental blood flow during alkalosis 
in the sheep. Amer J Physiol 228:1497-1500, 
1975. 

12. Makowski EL, Meschia G, Droegemueller W, 
Battaglia FC. Measurement of umbilical arterial 
blood flow to the sheep placenta and fetus in 
utero. Circ Res 23:623-631, 1968. 

13. Buckberg GD, Luck JC, Payne DB, Hoffman JIE, 
Archie JP, Fixler DE. Some sources of error in 
measuring regional blood flow with radioactive 
microspheres. J Appl Physiol 31:598-604, 1971. 

14. Rudolph AM, Heymann MA. The circulation of 
the fetus in utero: Methods for studying distribu- 
tion of blood flow, cardiac output and organ blood 
flow. Circ Res 21:163-184, 1967. 

15. Tuvemo, T. Role of prostaglandins, prostacyclin, 
and thromboxanes in the control of the 
umbilical-placental circulation. Semin Perinatol 
4:91-95, 1980. 

16. Tuvemo T, Strandberg K, Hamberg M. Contrac- 
tile action of a stable prostaglandin endoperoxide 
analogue on the human umbilical artery. Acta 
Physiol Scand 102:495-496, 1978. 

17. Weber PC, Larsson C, Anggard E, Hamberg M, 
Corey EJ, Nicolaou KC, Samuelsson B. Stimula- 
tion of renin release from rabbit renal cortex by 
arachidonic acid and prostaglandin endoperox- 
ides. Circ Res 39:868-874, 1976. 

Received April 12. 1982. P.S.E.B.M. 1982, Vol. 171. 



nuxKf.MN^A ^>f TH^ vAiF.rv r-'^fk hxrf^ttimtsjKL wcuxio amd 



m,! 



-55 I If 



The Sex-Related Drfterence in Perfluorooctanoale ExcrBtion in the Rat^ (4147IQ 
H. HANHUARVI,* R. H, QPHAUG,+ and L. SINGER+- 



"hi' parting nt of hrnuury. IJm\truty of Kmopio, FA>M. 138, SF-TOIOI Kuifpio 10. 
*ifi//f hi'miifr^ Frofcram. Srh//ol of l>enti3iry . Umhermy cfMmmeuHa, 515 Deiaware Street Sotakeast. 

MinneapolU, Mkmesaia 55455 



Ahstrari The ufiruiry excretion of pedhiofooctaiioic acid fPFOi tn i 
wiii^ invcMigiiicd. FemiOe rirtft excreted 76 ^ 2.7 (SEM^ of a 2'iiig dose of i 
(aft I^FO; in the urine in 24-1^ whereM male rau excreted oiil> 9.2 r 3.5<^ of the dose. TIk 
Ff-O clearance, inulin clearance, net excretion rMe of PFO. and the gkMiienilarfiltxation me 
i^ FH) were meaMired. The effect of probenecid, an inhMtor of the orpmic acid transport 
hyitUtn, on the%e meaturementft was abo detemined. In fennk rats the PPO clearance «a& 
MveraMold greater than the inulin clearance and the clrarance of PFO was markedly reduced 
by probenecid. Conversely, in male rats the PFO clearance was only a fraction of the inulin 
clearance and was virtually unaffected by probenecid. The dau indicate that female rats are 
able to rapidly eliminate PFO in the urine by an active secretoiy mechanism w-hich is 
inhibited by probenecid. In nude rats this secretory mechanism is either absem or relatively 
inactive, '/his difference in PFO excretion by the male and female may explain the sex- 
relaled difference in PFO toxicity in which male rats are more susceptible to high doses than 
females. 



(iuy i*t al. have reported the results of 
attemptfi in isolate and characterize the 
compound(s) comprising the nonionic 
fluorine fraction of human serum (1). They 
indicate that these compounds are mainly 
perfluoro fatty acid ((\-C\) derivatives and 
thill the major fluorocarbon isolated from 
human serum resembled perfluorooctanoic 
acid ((\l',ft(()()H). Since derivatives of 
peifluorooctanoatc (PFO) are widely used 
commeivially it is important to study its be- 
havjoi in livintt organisms. In 1980 Ophaug 
and Singer reported that PFO, administered 
by stnniacli intubation to female rats, was 
miuilly t'xcifted into the urine and by 96 hr 
ifiily iiiucs loiiuiined in the blood (2). Other 
invest inntnis have fed diets containing 
1000 /»/i||| 4,f ammonium perfluorooc- 

iii/I'^'"** "" overnight fast the serum of 
hno?*'"* '^'•^"^"incd 21 49 ppm of PFO 
... 1^"^ "*«►• s^fiiim of similarly treated 
^ ^^*^ contiiineil O.l.S 0.65 ppm. These 
"^ ^ontnnuHl ijk- ability of female rats 



• llUN NhuiN vK«s Mippo.u'.i hN Ciiani 1)1 OIKSOfiom 
*'*'*" .o,,OH,HMu|,,Ko Should K« acKlrcsscJ. 



to rapidly dispose of rather lai^ doses of 
PFO and illustrate a dramatic sex difference 
in the metabolic handling of this fluorocar- 
bon by rats. Janssen et al, have previously 
shown that there is also a sex-related differ- 
ence in the elimination rate of 1-amino- 
cyclohexanecarboxylic acid (ACHC) in 
rats (4). The principal aim of this investi- 
gation was to determine whether there is a 
sex-related difference in PFO excretion by 
the rat kidney. 

Materials and Methods. Holtzman rats 
fed rat chow (Purina) and tap water ad 
libitum were employed in all experiments. 

Four male and six female rats were ad- 
ministered 2 ml of an aqueous solution 
containing 2 mg of nonionic fluorine as PFO 
by stomach intubation. Seven female rats 
were administered 2 ml of distilled water as 
controls. The animals were then placed in 
individual metabolic cages and fed rat chow 
and tap water. A few crystals of thymol 
were added to the urine containers to in- 
hibit bacterial growth during the collection 
period. After 24 hr the animals were sac- 
rificed by cardiac puncture. The blood was 
allowed to clot and the serum was collected 
after centrifugation. The volume of the 
urine collections, including the volume of 






.V) 



ifwitmcnlai Hii»K>0\ diKt McUuinc 



SEX DIFFERENCE IN PFO EXCRETION IN RATS 



51 



water used to rinse the metabolic cages was 
recorded. The ionic fluoride content of the 
serum and urine was determined at pH S.O 
with the fluoride ion-specific electrode (S, 
6). The total fluorine content of the serum 
and urine was determined by the oxygen- 
bomb reverse extraction technique (7). The 
nonionic fluorine level of the serum and 
urine was calculated as the difference be- 
tween the ionic and total fluorine levels. 

For clearance studies of PFO and inulin 
the rats were anesthetized with Inactin 
(80-90 mg/kg ip), a barbituric acid deriva- 
tive which gives a 3- to 4-hr anesthesia. The 
femoral vein was cannulated for continuous 
infusion of 5% mannitol in isotonic saline 
and the femoral artery was cannulated for 
drawing blood samples. In order to obtain 
serial collections of urine the urinary blad- 
der was also cannulated. Intravenous (iv) 
priming doses of 5.2-5.6 mg of [1-'*C] 
ammonium perfluorooctanoate (sp act 
0.5 /LiCi/mg) and 8.8 fig of tritiated inulin 
(methoxy-^H, sp act 114 /LtCi/mg) were 
given to each animal. The radiolabeled in- 
ulin and PFO in 5% mannitol in isotonic 
saline was then infused at a rate of 0.21 
ml/min. An additional 0.42-0.63 mg/hr of 
^^C-labeled PFO and 9.6 /Ltg/hr of tritiated 
inulin was infused during the experiments. 
After a 45-min equilibration period the first 
plasma sample was collected for clearance 
calculations. Urine specimens were col- 
lected over 10-min intervals and additional 
arterial blood samples were obtained at the 
midpoint of each collection period. When 
the urine and serum collections for the 
clearance study were complete probenecid 
was administered (65-68 mg/kg ip) and, 
after 20-30 min, additional consecutive 
10-min clearance tests were performed to 
test the effects of probenecid on the organic 
acid transport system. 

In the cumulative excretion study the rats 
were prepared as described for the clear- 
ance tests except that arterial cannulation 
was not needed. The rats were dosed iv 
with a mixture of radiolabeled PFO (10- 
20%) and unlabeled PFO (80-90%). Five 

' (l-**C]Pcrfluorooctanoic acid (ammonium salt) was 
supplied by Minnesota Mining and Manufacturing 
Company, St. Paul, Minn. 55101. 



percent mannitol in isotonic saline was in- 
fused at a rate of 0.081 ml/min and urine 
specimens were collected over 30-min 
intervals. The effect of probenecid was as- 
sessed by administering 65-68 mg/kg ip at 
least 30 min prior to the administration 
of PFO. 

The blood samples were centrifuged and 
20 /Ltl of plasma was removed for radioac- 
tivity determinations. The rest of the 
plasma sample was transferred to Amicon 
Centriflo 2100 CF 50 ultrafiltration cones 
and centrifuged. These membranes retain 
molecules with a molecular weight greater 
than 50,000 daltons. Fifty microliters of the 
ultrafiltrate was also prepared for radioac- 
tive determination as was 50 jjlI of each 
urine sample. 

The ^H (inulin) and/or ^*C (PFO) activity 
in each sample was counted in a dual- 
channel Packard Tri-Carb liquid scintilla- 
tion spectrometer. A series of quenched 
standards and the pH]inulin and [**C]PFO 
standards were also counted. The count 
rates obtained for all samples were cor- 
rected for background, quenching (by use 
of the automatic external standard method), 
^H spillover into the **C counting channel, 
and **C spillover in the ^H counting chan- 
nel. All samples were counted until 20,000 
counts had been recorded. 

The clearances (CO were calculated 
using 



Cl (ml/min/100 g) = 



UxV 
P X W' 



where 

U = urinary concentration of the com- 
pound (/ig/ml), 

V = urine flow rate (ml/min), 

P = plasma concentration of the un- 
bound compound (/ig/ml), 

W = animal weight (g)/100. 

The net excretion rate (NE) was calculated 
(/ig/min/100 g) using 



NE = 



UxV 

w 



52 



SEX DIFFERENCE IN PFO EXCRETION IN RATS 



where 

U = urinary concentration of the com- 
pound (/ig/ml), 
V = urine flow rate (ml/min), 
W = animal weight (gVlOO. 

The cumulative excretion percentage was 
calculated as 



where 

U = urinary concentration of the com- 
pound (/ig/ml), 
VF = total urine volume (ml), 
P = total dose of the compound (/ig). 

The glomerular filtration rate (F) of PFO 
was obtained from 

F (/ig/min/100 g) = P x C,. (inuiin>^ 

where 

P = plasma concentration of the un- 
bound compound (/ig/ml), 
Ci - inulin clearance (ml/min/100 g). 

All the tests were performed at least twice 
unless otherwise stated. 

Results. The ionic and nonionic fluorine 
levels of the scrum and the percentage of 
the dose of nonionic fluorine excreted in the 
urine 24 hr after administration of 2 mg of 
nonionic fluorine, as PFO, to male and 
female rats are presented in Table I. 



Twenty-four hours after admmistiatioo d 
the dose, female rats had excreted 76 ± 
2.7% of the dose in the urine and had a 
mean serum nonionic fluorine level of 0.35 
ppm. Although this serum nomonic fluorine 
level is considerably higher than that ob- 
served in female rats which did not receive 
a dose of nonionic fluorine (0.07 ± 0.02 
ppm), male rats had serum nonionic 
fluorine levels which were much higher 
(44.0 ± 1.7 ppm) and had excreted only 9.2 
± 3.5% of the dose of nonionic fluorine in 
the urine. The serum ionic fluoride levels of 
male and female rats given PFO were not 
different from that of undosed female rats. 
PFO was bound to a similar extent in the 
plasma of male and female rats. A mean of 
97.5 ± 0.25% (SEM, N = 16) of the PFO in 
the plasma was bound. The results of clear- 
ance studies on male and female rats are 
shown in Table II. In repeated tests it be- 
came obvious that there was a crucial dif- 
ference in PFO clearance and the PFO/ 
inulin clearance ratio between sexes. The 
PFO clearance in female rats was several 
times greater than the inulin clearance. The 
administration of probenecid (65-68 mg/kg 
ip), which strongly inhibits the renal active 
secretion system for organic acids (9), dra- 
matically reduced the clearance ratio in 
female rats. The net excretion of PFO was 
reduced from 4.6 to 0.13 /ng/min/lOO g fol- 
lowing the administration of probenecid. In 
male rats, however, the PFO/inulin clear- 
ance ratio and the net excretion of PFO 
were virtually unaffected by probenecid. 



lAHM-: I. Si RUM Ionic Fluoride and Nonionic Fluorine Levels and Urinary 

i:x< Ri HON OF Nonionic Fluorine 24 hr after Administration" of a 2-mg Dose of 

Nonionic Fluorine as PFO to Mai e and Female Rats 





lYcatmcnt 
No dose 
IX>sed 
IX>sed 




Serum 


Percentage of 

dose excreted 

in urine 


Sex 


Ionic fluoride 
(ppm) 

0.032 ± 0.003'* 

(7) 
0.020 - 0.003 

(6) 
0.033 *: 0.003 

(4) 


Nonionic fluorine 
(ppm) 


It male 
leiniile 
Male 


0.07 ± 0.02 

(5) 
0.35 ±0.11 

(6) 
44.0 ± 1.7 

(4) 


76 ± 2.7 

(4) 
9.2 ± 3.5 

(4) 


" (iastric i 
" Mean * 


intubation 
SFM (No. of animals). 









SEX DIFFERENCE IN PFO EXCRETION IN RATS 



53 



TABLE II. Effect of Probenecid (65-68 mg/kg ip) on the PFO/Inulin Clearance 
Ratio, Net PFO Excretion, and Filtered PFO in Male and Female Rats 



Female 



Male 



No probenecid After probenecid No probenecid After probenecid 

PFO clearance 5.8" (5.5*-6.0) 0.11 (0.08-0.14) 0.17 (0.16-0.17) 0.10 (0.09-0.11) 

(ml/min/lOO g) 
PFO clearance 14.5 (8.2-20.7) 0.46 (0.43-0.48) 0.22 (0.17-0.26) 0.12 (0.11-0.13) 

inulin clearance 
Net PFO excretion 4.6 (3.9-5.2) 0.13 (0.09-0.17) 0.17 (0.13-0.21) 0.12 (0.11-0.12) 

(jig/mm/mg) 
Filtered fraction 0.42 (0.20-0.64) 0.29 (0.21-0.37) 0.75 (0.72-0.77) 1.00 (0.91-1.10) 

(Mg/min/lOO g) 

" Mean (range of values) for two animals. 

^ Mean values for each animal of two to four determinations. 



Table III presents the data obtained in 
cumulative excretion studies of PFO in the 
urine of male and female rats over a 7-hr 
period. Female rats were observed to ex- 
crete 76% of the administered PFO (23-25 
mg/kg) whereas male rats excreted only 
7.8% of the dose. Intraperitoneal injection 
of probenecid (65-68 mg/kg) given at least 
30 min before administration of the PFO 
modified the cumulative excretion curve for 
males only slightly. In female rats, how- 
ever, probenecid markedly reduced PFO 
elimination to 11.8%. 

Discussion. Griffith and Long have 
clearly shown a sex-related difference in 
PFO toxicity in rats being fed PFO in their 
diet (3). The liver appeared to be the target 
organ in rats, and males were found to be 
more susceptible to high doses of PFO than 
females. The PFO concentration in pooled 
plasma and liver specimens was consid- 



erably higher in male rats than in similarly 
treated females. 

Twenty-four hours after receiving a 2-mg 
dose of nonionic fluorine (as PFO) male rats 
had serum nonionic fluorine levels that 
were more than 100-fold higher than that of 
similarly treated females (Table I). The 
serum ionic fluoride levels of male and 
female rats were not significantly increased 
following administration of the PFO. This 
provides good evidence for the metabolic 
stability of PFO in rats. The conclusion that 
PFO is metabolically stable in rats is sup- 
ported by the demonstration of quantitative 
recovery of nonionic fluorine in the urine 
and feces of female rats given PFO (2). In 
addition, Hagen et al. have recently dem- 
onstrated the accumulation of PFO in the 
serum of male rats given a single oral dose 
of lH,lH,2H,2H-perfluorodecanol (8). In 
contrast to female rats which excreted 76 ± 



TABLE III. Effect of Probenecid (65-68 mg/kg ip) on the Cumulative Urinary 
Excretion of PFO in Male and Female Rats" 





Female 






Male 


Hours 


No probenecid 


After probenecid 


No probenecid 


After probenecid 


1 
2 
3 
4 
5 
6 
7 


21.6(20.2-22.9)° 
36.3 (32.7-39.9) 

46.1 (38.6-53.5) 
56.5 (49.5-63.4) 
65.7 (59.3-72.0) 
71.9(64.8-78.9) 

76.2 (68.9-83.5) 


1.1 (1.1-1.1) 
2.5 (2.2-2.7) 
3.6(3.1-4.1) 
5.3 (4.0-6.6) 

7.2 (4.8-9.6) 
9.2(5.8-12.6) 

11.8(6.6-17.0) 


1.4(0.9-1.8) 
2.4(1.6-3.1) 
3.4(2.4-4.3) 
4.5 (3.5-5.4) 

5.7 (4.9-6.4) 

6.8 (6.3-7.3) 
7.8(7.3-8.2) 


1.0(0.9-1.1) 
1.7(1.5-1.8) 
2.4 (2.2-2.6) 
3.1 (2.8-3.3) 
3.7 (3.3-4.0) 
4.4 (3.9-4.8) 
5.0 (4.4-5.5) 



' Results are percentage of dose excreted. 
^ Mean (range of values) for two animals. 



54 



SEX DIFFERENCE IN PFO EXCRETION IN RATS 



2.7% of the PFO dose in the urine after 24 
hr, male rats excreted only 9.2 ± 3.5% of 
the dose (Table I). These data indicate that 
the sex-related difference in PFO toxicity in 
rats (3) is due to the relatively slow urinary 
excretion of PFO in male rats. 

Since inulin is excreted only by glo- 
merular filtration and not actively se- 
creted in renal tubuli, the observation of 
PFO/inulin clearance ratios that were sub- 
stantially greater than 1.0 for female rats 
(Table II) provides evidence that, in the 
female rat, PFO is excreted in part by an 
active secretion mechanism. The fact that 
probenecid rapidly decreases the PFO/ 
inulin clearance ratio from 14.5 to 0.46 
strongly supports this conclusion. Since the 
PFO/inulin clearance ratio for male rats was 
less than 1 .0 and not significantly altered by 
the administration of probenecid it appears 
that active tubular secretion of PFO in 
males either does not occur or occurs at an 
insignificant rate. Additionally, both the 
female rat after receiving probenecid and 
the male rat throughout the clearance 
studies had lower PFO than inulin clear- 
ance. This indicates that there is partial 
tubular reabsorption of PFO in both sexes. 
Janssen et al. have shown that there is a 
sex-related difference in the tubular reab- 
sorption of 1-aminocyclohexanecarboxylic 
acid (ACHC) in the rat kidney (4). ACHC is 
not bound to plasma proteins whereas in 
the present study 97.5 ± 0.25% (SEM) of 
the PFO was bound. Thus the excretion of 
ACHC by glomerular filtration occurs to a 
much greater extent than PFO. Our data 
indicate that there is a sex-related differ- 
ence in the active secretion of PFO. It is 
therefore possible that female rats also ac- 
tively secrete ACHC to a greater extent 
than male rats. 

The cumulative urinary excretion data in 
Table III illustrates the striking sex-related 
difference in PFO elimination. At doses of 
23-25 mg/kg male rats, because of a limited 
or completely inactive secretion mecha- 
nism, are able to net-excrete in 7 hr only 
about 10% of the amount of PFO excreted 
by females. Although probenecid had little 
effect on the cumulation excretion curve for 
males a low level of active tubular secretion 



of PFO cannot be ruled out. It is possible 
that probenecid, as an acid, slightly in- 
creases the tubular reabsorption of PFO 
and partially masks the active secretion. 

Ophaug and Singer found that female rats 
excreted 61% of the administered dose of 
nonionic fluorine (PFO) in 8 hr (2). The 
more rapid excretion (76% in 7 hr) observed 
for female rats in the cumulative excretion 
studies (Table III) is probably due to the 
fact that the animals in the present study 
were infused with 5% mannitol in isotonic 
saline during the excretion period and that 
the PFO was administered iv rather than by 
gastric intubation. After the administration 
of probenecid to female rats the rate of 
excretion of PFO decreased dramatically 
and after 1 hr there was no difference in 
PFO elimination between sexes. Griffith 
and Lx)ng observed that rhesus monkeys do 
not exhibit a sex-related difference in the 
elimination of PFO (3). Serum PFO levels 
(following an overnight fast) ranged from 45 
to 71 ppm for males receiving 3 or 10 mg 
PFO/kg/day as compared to levels of 50 to 
79 ppm for corresponding females. The high 
serum levels of PFO found in both sexes of 
rhesus monkey closely reflect the situation 
found in male rats in that they appear to 
eliminate a dose of PFO rather slowly. One 
might speculate, therefore that the pro- 
benecid-sensitive active secretory system 
we have observed in female rats is absent 
or inactive in both sexes of the rhesus 
monkey. 

Based upon these data it is concluded 
that the female rat possesses an active se- 
cretory mechanism which rapidly elimi- 
nated PFO from the body. This secretory 
mechanism is lacking or relatively inactive 
in male rats and accounts for the greater 
toxicity of PFO in male rats. 



1. Guy WS, Taves DR, Brey WS. Organic 
fluorocompounds in human plasma: Prevalence 
and characterization. In: Filler R, ed. Biochemis- 
try Involving Carbon -Fluorine Bonds. ACS Sym- 
posium Series No. 28. Washington. DC, Amer 
Chem Soc, Vol 7:pll7, 1976. 

2. Ophaug RH, Singer L. Metabolic handling of 
perfluorooctanoic acid in rats. Proc Soc Exp Biol 
Med 163:19-23, 1980. 



SEX DIFFERENCE IN PFO EXCRETION IN RATS 



55 



3. Griffith FD, Long JE. Animal toxicity studies with 
ammonium perfluorooctanoate. Amer Ind Hyg 
Assoc J 41:576-583, 1980. 

4. Janssen FW, Young EM, Ruelius HW. Effect of 
sex hormones on the disposition in rats of I- 
aminocyclohexane carboxylic acid, a metabolite 
of a semisynthetic penicillin. Drug Me tab Dispos 
4:540-546, 1976. 

5. Singer L, Armstrong WD. Determination of fluo- 
ride in ultrafiltrates of sera. Biochem Med 
8:415-422, 1973. 

6. Singer L, Armstrong WD, Vogel JJ. Determination 
of fluoride content of urine by electrode potential 
measurements. J Lab Clin Med 74:354-458, 1969. 

7. Venkateswariu P. Determination of total fluorine 



in serum and other biological materials by oxygen 
bomb and reverse extraction techniques. Anal 
Biochem 68:512-521, 1975. 

8. Hagen DF, Belisle J, Johnson JD, Venkateswariu 
P. Characterization of fluoridated metabolites by 
a gas chromatographic-helium microwave plasma 
detector— The biotransformation of 1H,1H,2H,2H- 
perfluorodecanol to perfluorooctonate. Anal Bio- 
chem 118:336-343, 1981. 

9. Weiner IM, Mudge GH. Renal tubular mecha- 
nisms for excretion of organic acids and bases. 
Amer J Med 36:743-762, 1964. 



Received January 8, 1982. P.S.E.B.M. 1982, Vol. 171. 



PWtCPPt^SCA OP THP SOriRTY POR eXPRRlMENTAL BIOLOGY AND MEDICINE I7t» 56— M ( I98Z) 



Localization of Radiolabeled Antibody in SVT2 Tumor Increases with 
Immunosuppression of the Host (4t477) 

DONALD J. BUCHSBAUM,' JANET M. ANDERSON, and BRUCE E. BRAY 

f>fpaftment of Therapeutic RadUUoity. Box 494. University Hospitais, University of Muutesotu. 

.WinneapfUis. .Minnesota 55455 



Abstract. The locaiization of radiolabeled tumor-ipecific antibodies to an SV40^ 
transformed mouse rumor was analyzed in inuniino<iuppnB9.ied (X-ray and cortiaone) and 
nonimmunosuppressed mice. (C57BKS ^ Bailyc)F, mice were immunized with die SVT3 
tun(W>r of Balb/c origin. Radiolabeled antibody was isolated from '^[4abeied imnnme gamma 
globulin by adsorption onto $VT2 cells, and elution from these cells using citrate bii£kr. The 
radiolabeled antibodies were injected into normal (C57B16 ^ BalbciF, mice. The purified 
antibodies present in this serum bound specifically in vitro to SV40-transfornied cell lines, in 
vivo, the '"Mabeled antibodies localized preferentially in the SVT2 tumor in inummosup- 
pressed mice. Significantly less "^I-labeled antibody localized in the SVT2 tumor in noaiufr- 
munosuppressed mice. The localization of '^I-labeled antibody in SVTl tumor in im- 
munosuppressed mice was reduced significantly by passive administration of anti-SVT2 
serum. 



Cancer diagno^i^ or therapy using a 
tumor localizing radiolabeled antibody has 
been the goal of several groups of inves- 
tigators (1-7). Most of the work has been 
done in xenogeneic systems, which require 
exhaustive absorptions of immune serum 
with normal cells to obtain a more tumor- 
specific serum. It would be desirable to 
work in a syngeneic system in which the 
only antibody response elicited is that 
against unique antigens on the tumor cell 
surface. Simian virus 40 (SV40)-trans- 
formcd mouse cells possess an antigen 
specific for SV40-induccd tumors that is 
absent in non-SV40-induccd tumors (8~ 
II). It has been shown that immunization 
of (C57BI/6 X Balb/c)F, mice with syn- 
geneic SV40 tumor (SVT2) elicits anti- 
bodies reacting against the SV40 tumor- 
associated cell surface antigen (10, 11). 
Wc chose this syngeneic system to avoid 
the cross-reactivity problems often asso- 
ciate(l with the use of tumor localizing 
antibodies prepared from xenogeneic sc- 
niin (2). 



' In vvimm \ oMi's|nnuliMKi* sluMiUI Iv juKlicsscd. 



Two possible limitations in the use of 
radiolabeled antibodies are the presence of 
circulating tumor antigens and host anti- 
body. Tumor antigens which are shed finooi 
the solid tumor may combine with in- 
jected radiolabeled antibody, and the com- 
plexes cleared. However, tumor localiza- 
tion by radiolabeled antibodies has oc- 
curred even in the presence of high levels of 
circulating CEA (4, 5, 12). Antibodies pro- 
duced by the host to its growing tumor may 
bind to the tumor, masking the binding sites 
from the labeled antibody (13, 14). Here we 
report that immunosuppression of the host 
prior to transplantation of an SV40-trans- 
formed mouse tumor led to significantly 
increased tumor localization of radio- 
labeled tumor-specific antibodies. This 
tumor localization was blocked by passive 
administration of immune serum. 

Materials and Methods. Mice and im- 
munizxition. Eight- week-old male (C57B1/6 
X Balb/c)F, mice (Animal Genetics and 
Production Branch, National Cancer Insti- 
tute, Frederick, Md.) genetically identical 
to those used by Ting (10, 11) were im- 
munized with the SVT2 tumor. The SVT2 
and SVA3I C 14 lines of Balb/c embryo cells 



» iM*\ti|il<* ' I'^HMn ihi- SiHirf\ lot rxiH'nmrniiil Hitilonv ami Mrdictnr. 
Ml ii|ihi« i«-k»*i\i>il 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



57 



transformed by SV40 were obtained frozen 
from a tumor bank (Biotech Research 
Laboratories, Inc., Rockville, Md.) rec- 
ommended by Dr. C. C. Ting at the NCI. 
They were grown as monolayers in 150-cm^ 
tissue culture flasks (Coming Glass Works, 
Corning, N.Y.) in RPMI 1640 medium 
containing 5% heat-inactivated calf serum 
(Grand Island Biological Co., Grand Island, 
N.Y.), 100 units/ml penicillin, 100 fig/ml 
streptomycin, and 2 mM L-glutamine at 
37° in a humidified atmosphere with 5% 
C02-95% air. The cells were harvested by 
exposure to 0.1% trypsin- 10 mAf EDTA in 
PBS, pH 7.2. The mice were immunized 
with the SVT2 tumor by inoculation of 1 x 
10^ cells into the footpad. Antisera were 
collected by cardiac puncture from mice 
with progressively growing tumors 4 to 6 
weeks after tumor inoculation. 

Isotopic antiglobulin technique. Sera 
were assayed for antibody activity by an 
isotopic antiglobulin technique (15). The 
antiglobulin used in this procedure was the 
IgG fraction of sheep anti-mouse IgG (Cap- 
pel Laboratories, Cochranville, Pa.) radio- 
labeled with '2*1 (Amersham Corp., Chi- 
cago, 111.) at a specific activity of 0.25 
mCi/mg by the iodine monochloride method 
(16). Briefly, 5 x 10* target cells in 0.1 ml 
HBSS were incubated first with 0.1 ml of a 
1:50 dilution of antiserum or normal serum 
for 30 min at room temperature. The cells 
were washed four times with 1 ml HBSS 
containing 10% -y-globulin-free newborn 
calf serum (Grand Island Biological Co., 
Grand Island, N.Y.). A second incubation 
with 10 /Ltl '^I-labeled sheep anti-mouse Ig 
containing approximately 3 x 10® cpm was 
done at 0"* for 15 min, and the cells were 
washed seven times. All tests were done in 
duplicate. The variation among duplicates 
did not exceed 10%. Activities of the anti- 
sera are expressed as an absorption ra- 
tio (AR): 



AR = 



'2*1 cpm on cells incubated with 

mouse antiserum 
-f '25i.antiglobulin 

'2*1 cpm on cells incubated with * 

normal mouse serum 

-f '2si-antiglobulin 



A reaction was considered positive when 
the AR was ^3. 

Gamma globulin preparation. Gamma 
globulin was prepared from pooled anti- 
SVT2 antiserum by three precipitations 
with neutral 50% saturated ammonium sul- 
fate. The final precipitate was dissolved in a 
volume of PBS equal to 40% of the original 
serum volume. Normal gamma globulin 
was prepared by ammonium sulfate pre- 
cipitation of (C57B1/6 X Balb/c)F, serum; 
the final precipitate was dissolved in a vol- 
ume of PBS equal to 27% of original serum 
volume. 

Antibody purification and radiolabeling. 
The immune gamma globulin was enriched 
before labeling by affinity purification using 
a modified version of the procedure of 
Ehrlich and Witz (17). SVT2 cells were 
washed twice with PBS, and resuspended 
in serum-fi-ee RPMI 1640 medium at a con- 
centration of 2.2 X 10^ cells/ml. Aliquots 
containing 1.35 ml of the SVT2 cell suspen- 
sion and 0.15 ml anti-SVT2 serum were 
combined in 15-ml conical test tubes (Fal- 
con Plastics, Los Angeles, Calif.). These 
were incubated in an ice bath for 1 hr with 
occasional shaking. After incubation, the 
cells were pelleted and washed three times 
with PBS. Antibody remaining adsorbed to 
the packed cells was eluted by resuspend- 
ing each pellet in 0. 1 ml of 0. 1 Af citrate 
buffer, pH 3.5, for 15 min. The supema- 
tants were pooled, neutralized with 1 N 
NaOH, and dialyzed against one liter of 
PBS overnight. The entire procedure — 
absorption, washes, elution, and dialysis — 
was carried out at 4"*. The dialyzed eluate 
was added to 1.5 mg of anti-SVT2 Ig to 
produce "enriched" immune Ig. 

Two preparations of enriched immune Ig 
were radiolabeled by the iodine mono- 
chloride method (16). Gamma globulin, 
1.5 mg, was dialyzed against one liter of 
borate buffer (0.20 M boric acid, 0.16 M 
NaCl, 0.04 M NaOH, pH 8.0) overnight 
before iodination. '2*1 (15 mCi) and 4 eq of 
ICl were used to label the immune gamma 
globulins to a specific activity of 4.5 and 5.3 
mCi/mg. Normal (C57B1/6 x Balb/c)Fi 
serum was used as a protective protein for 
the labeled preparations at 20% of the final 



58 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



volume. One milligram of normal Ig was 
labeled in a similar fashion with 1 mCi of 
'^'Iand4e4ofICl. 

The radiolabeled antibody preparations 
were once afTinity-purified by the procedure 
described above. Aliquots containing 2.25 
ml of the SVT2 cell suspension and 0.25 ml 
'^M-labeled enriched anti-SVT2 Ig were 
combined and incubated in an ice bath 1 hr. 
Washes, elution, and dialysis have been 
detailed above. The '^'^Mabeled once affin- 
ity-purified antibodies were further purified 
by In vivo screening (18). Approximately 
400,000 and 700,000 cpm of Preparations I 
and 11 of once-absorbed and eluted anti- 
body in 0.5 ml PBS were each ii\jected 
ip into four normal mice. The mice were 
bled 24 hr later by cardiac puncture. Ap- 
proximately 15% of the ii\jected counts 
were recovered in the serum. '^'I-labeled 
normal Ig was also In vivo screened. Ap- 
proximately 1.5 X I0« cpm of >»4-labeled 
normal Ig was if\iected ip into five normal 
(C57B1/6 X Balb/c)F, mice. Approximately 
3% of the counts ii\jected were recovered in 
the serum 24 hr later. 

A portion of the enriched immune gamma 
globulin was passed over a protein A- 
sepharose column (19), and the individual 
IgO fractions were radiolabeled with '^^I. 
Each fraction was affinity-purified with 
SVT2 cells and in vivo screened in normal 
(C57BI/6 X Balb/c)F, mice, and tested for 
binding to finely minced tumor pieces taken 
from immunosuppressed and nonimmuno- 
supprcsed mice. 

In vivo distribution studies. The in vivo 
distribution of the radiolabeled purified 
anti-SVT2 antibodies was studied in non- 
immunosuppressed and immunosuppressed 
mice. The immunosuppressed mice con- 
sisted of a gn>up of 1 1 (C57B1/6 x BalWc)F, 
mice treated with 400 R of whole-body ra- 
diation from a General Electric 220-keV 
Maximar X-ray machine the day before 
tumor transplantation^ and given a 0.25 mg 
sc dose of coilis^>nc acetate on alternate 
days beginning with the day of tumor trans- 
plant. SVT2 tumors weix started in the 
right tlank in these mice with trvKar trans- 
plants of SV T2 tumor maintained in irradi- 
ated iC5'Bl 6 \ Balbc>F, mice. MOPC-315 



tumors were started in the left flank in these 
mice with trocar transplants of MOPC-31S 
tumor maintained in Balb/c AnN (The 
Charles River Breeding Laboratories, Inc., 
Wilmington, Mass.) mice. The five nonim- 
munosuppressed mice received no corti- 
sone or X-ray. The tumors were started in 
these mice at the same time they were 
transplanted into the immunosuppressed 
mice. Nineteen days after transplant, four 
tumor-bearing immunosuppressed mice re- 
ceived 0.3 ml of unlabeled (C57B1/6 x 
Balb/c)Fi anti-SVT2 serum ip. The next 
day, these four mice (Group III) and seven 
other immunosuppressed mice that were 
not given unlabeled antiserum (Group II) 
received ip a mixture of 0.01 fxCi of Prep- 
aration I of '^*I-labeled enriched once af- 
finity-purified/iVf v/vo-screened antibody 
and 0.02 /nCi *^4-labeled, in v/vo-screened 
normal (C57B1/6 x Balb/c)Fi gamma globu- 
lin. Group I containing five nonimmuno- 
suppressed mice received 0.01 /LtCi of Prep- 
aration II of *"I-labeled enriched once 
affinity-purified//n vivo screened antibody 
and 0.02 /LtCi **4-labeled, in vivo screened 
normal gamma globulin. Forty-eight hours 
after injection, mice were bled by cardiac 
puncture, dissected, and the tissues counted. 
For a week prior to injection of radiolabeled 
antibody and until dissection, the mice 
were given water containing 6 x 10"* Af KI 
to block thyroid uptake of iodine that could 
be freed by the catabolism of the labeled 
gamma globulins. 

Results. High absorption ratios against 
SVT2 cells were obtained when testing 
serum from all groups of immunized mice. 
The sera were pooled, and the effect of 
concentration of antiserum on the absorp- 
tion ratio obtained with SVT2 tumor cells 
indicated a titer of 1:8192. Table I demon- 
strates that the pooled anti-SVT antiserum 
reacted with SVT2 and SVA31 C14 cells, 
but did not bind to MOPC-315 myeloma 
cells of Balb c origin, P-815-X2 mas- 
tocytoma cells of DB.\ 2 origin, RIF-1 
radiation-induced fibrosarcoma cells of 
C3H HeJ origin, B16-FI melanoma cells of 
C57BI 6 origin, or SCK spontaneous tumor 
cells of .\ J origin. 

The specificit\ of the ^^'^Mabeled anti- 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



59 



TABLE I. Absorption Ratios of (C57BLy6 x 

Balb/c)F, ANTI-SVT2 Antiserum 

WITH Various Tumor Cells 



TABLE II. In Vitro Binding of '^M-Labeled 

(C57BLy6 X Balb/c)F, Anti-SVT2 
Antibody at Various Stages of Purification 









cpm of 






Binding (%) to 




Target 


cpm of 


normal 






tumor cells" 


Scrum" 


cells 


antiserum 


serum* 


AR*^ 


















Purification stage 


SVT2 


P-815.X2 


Pdoled 


SVT2 


27,% 1 


1157 


24.2 








progressor 


SVA31 C14 


18,327 


1230 


14.9 


1. Enriched '^I-gamma 
globulin 








MOPC-315 


5,113 


4096 


1.3 








P-815-X2 


1.192 


1268 


0.9 


Preparation I 


1.2 


— 




RIF-1 


1,141 


1018 


1.1 


Preparation II 


0.8 


0.8 




B16-F1 
SCK 


1.281 
1,680 


984 
1131 


1.3 
1.5 


2. Enriched, once absorbed 
and eluted 
Preparation I 


42.0 




■ A 1:50 dilution of mouse serum was incubated with 5 x 10* 


6.7 


target ceDs for 30 min at room temperature. The cells were 


Preparation II 


43.7 


3.5 


then washed four times, incubated with 10 ^1 of the ''^I-anti- 


3. Enriched, once absorbed 
and eluted/m vivo 






globulin for 15 min at 0", and then washed seven times. Each 
determination was performed in duplicate. 






* Control serum was normal (C57B1/6 x 


Balb/c)F, atal:50 


screened 






dilution. 










Preparation I 


37.2 


0.2 


'^ Absorption 


ratio = '"I . 


cpm antiserum/"*! cpm 


normal 


Preparation II 


46.2 


0.3 


serum (average of duplicate samples). 






4. Enriched, twice absorbed 
















and eluted 
















Preparation I 


54.3 


3.3 



SVT2 antibody was tested at various stages 
of purification by an in vitro binding assay, 
as detailed in Table II legend. Table II 
shows the specific binding of enriched, 
affinity-purified, and affinity-purified/iAi 
viVo-screened antibody preparations. The 
affinity purification brought the specific 
binding to SVT2 cells fi-om 1.2 to 42.0- 
43.7%. In vivo screening of affinity-puri- 
fied antibodies did not increase the binding 
to SVT2 cells, but did reduce the nonspe- 
cific binding to P-815-X2 mastocytoma cells 
essentially to zero. 

The dose -response effect of immuno- 
suppression and change in titer of circu- 
lating host antibody to SVT2 is shown in 
Table III. We used techniques of analysis of 
variance and regression to determine the 
affect of radiation and cortisone on the AR. 
There is a significant slope (linear effect) of 
radiation on the absorption ratio for both 
the with- and without-cortisone groups of 
mice. However, the two slopes are signifi- 
cantly different with the steeper slope as- 
sociated with the cortisone group. 

The tissue to blood ratios of *^I-labeled 
anti-SVT2 antibody and *^4-labeled normal 
gamma globulin in the three groups of 
tumor-bearing mice arc shown in Table IV. 
The SVT2 tumor load (mean ± SE) in 



" SVT2 tumor cells or P-8I5-X2 control cells were 
suspended in RPMI 1640 medium containing 5% calf 
serum at a concentration of 1.7 x 10' cells/ml. Six- 
tenths ml of the cell suspension (10' cells) was placed 
into 12 X 75-mm glass tubes, and a volume of antibody 
containing a minimum of 3000 cpm was added to 
duplicate tubes. The tubes were incubated at 3T for 
1 hr with occasional shaking, and then counted in a 
Beckman Gamma 7000 counting system to determine 
the total cpm added. After three washes with 1 ml of 
RPMI 1640 medium containing 10% calf serum, the 
tubes were counted again. Mean values are shown. 
The variation between duplicate tubes did not ex- 
ceed 10%. 

TABLE III. Effect of Immunosuppression 
ON Change in Titer of Circulating Host 
Antibody in Mice Bearing SVT2 Tumors 





No. of 






mice m 




Treatment 


group 


AR° 


OR 


3 


5.1 ±0.4 


R and cortisone 


4 


5.2 ± 0.4 


200 R 


4 


4.0 ± 0.8 


200 R and cortisone 


3 


3.6 ± 0.3 


400R 


4 


3.7 ± 0.5 


400 R and cortisone 


6 


1.6 ±0.4 



° Mean ± SE obtained by dividing counts of '**I- 
labeled sheep anti-mouse Ig bound to SVT2 cells in- 
cubated with serum obtained from individual mice in 
the diflferent treatment groups, 3 weeks after sc trans- 
plantation of the SVT2 tumor, by the counts bound 
to SVT2 cells incubated with normal serum. 



m 



tanwrnn ux:AUZAwm» a wrzm 



z 



41 



jr ff «r *^. — — *i 



2:551 



I 



^ I 



/> 


^ 


5«C 


^ 


•^ 


/• 






Z 


i 


< 








s 


/ 


«r 


< 


5 


f» 


^ 


m» 




> 


*" 


VJ 


^ 




^ 


< 


2 



::2 

£ I 

"-6 



fiQ 
< 



X ^3£32S5^ 



=1 
II 



2; 

E 



< K '• ir s X s Y s ar. fe 



_. 5 ^. r» ^ — *-; «r. r- ^'i r- — 
JT C ^, «r, — r>» r>« -^r r>f «r, — — 



Z oooooooooo 

< r* < — ^ ^ ^ — 50 O ae 
J, — « — ^* #< ^ #*; V <*v o 



g£ doooo 00000 

CO d ©' o' d d o' o' o' d o' 

^ ^ , tf tt 4-1 +1 f ■ +i +■ ♦ i> *! 

::" e » »?;**% — — n ^ n — ^ — 

% i 



= 5 « « 5 






1^^^ 



— — ooooo© 
ddddddddd© 

+1 +1 4-1 +1 fl +1 fl +1 +1 +1 

r**. f*\ 00 f*\ r* ^ 00 rj — tr-, 
•r, f; — ri ri r^. ri — •r-, — 

dddddddddd 



2 ' 



s 



>,^ .= 






iiU 

Illi 
= 111 

- ra '13 * 
>^c c a 

3 S c -^ 

-3 "O 

> JL 






o o 



— ^ 

1/3 
3 « 



-3 "3 . 

' C/3 S ^ 

'*! 

00 









ANTIBODY LOCALIZATION IN SVT2 TUMOR 



61 



Groups I-III was 0.31 ± 0.12, 0.44 ± 0.20, 
and 0.36 ± 0.11 g, respectively. The differ- 
ence in SVT2 tumor load between any of 
the groups was statistically insignificant. 
There was significantly less ^^'^I-labeled 
antibody localized in the SVT2 tumor in 
Group I of nonimmunosuppressed mice 
compared with Group II of immunosup- 
pressed mice, determined by two sample t 
test (P < 0.01). The localization of ^un- 
labeled normal gamma globulin was lower 
in SVT2 tumor and skin in the nonim- 
munosuppressed mice (Group I, P < 0.01). 
The localization of *^*I-labcled antibody in 
SVT2 tumor in immunosuppressed mice 
(Group II) was reduced by passive admin- 
istration of 0.3 ml anti-SVT2 serum (Group 
III, P < 0.01). The localization of ^"i- 
labeled antibody and ^^4-labeled normal 
ganuna globulin was lower in the MOPC- 
315 tumor in mice injected with anti-SVT2 
serum (Group III) compared with im- 
munosuppressed mice (Group II, P < 0.01), 
but was similar to that seen in nonim- 
munosuppressed mice (Group I). The sera 
from the individual animals in the three 
groups were tested by the isotopic anti- 
globulin technique. The mean values for 
Groups I-III were 6.6, 1.8, and 5.6, re- 
spectively, indicating the presence of host- 
anti-SVT2 antibody in the nonimmunosup- 
pressed mice. Table V shows the in vitro 
binding of the radiolabeled IgG antibody 
fractions to crude cellular preparations 



made by finely mincing SVT2 tumors taken 
from immunosuppressed and nonimmuno- 
suppressed mice. There was lower binding 
to tumors taken from nonimmunosup- 
pressed mice. Exposure of the cellular 
preparation made from tumor grown in 
nonimmunosuppressed mice to citrate 
buffer, pH 3.5, did not result in an eluate 
that showed antibody activity by the iso- 
topic antiglobulin technique. Incubation 
of the cellular preparation with ^^I-labeled 
sheep anti-mouse IgG did not result in 
greater binding of the antiglobulin than that 
seen with a cellular preparation made from 
tumors grown in immunosuppressed mice. 
Discussion. The possibility of using 
radiolabeled antibodies in syngeneic tumor 
models has been investigated in several 
other laboratories. Witz et aL (20) im- 
munized syngeneic mice with Moloney 
virus-induced lymphomas. When labeled 
affinity-purified antibodies were injected 
into animals bearing the tumor, essentially 
the only organ which showed preferential 
uptake of the labeled antibody was the 
spleen. There was no such spleen fixation 
in normal mice, suggesting that the spleens 
from tumor-bearing mice contained tumor 
cells, antigen, or virus. De Vaux Saint Cyr 
(21) found no tumor localization when 
studying a radiolabeled syngeneic antibody 
against a hamster SV40 tumor. This lack of 
localization may have been the result of a 
disappearance of the virus-induced antigens 



TABLE V. In Vitro Binding of Anti-SVT2 Immunoglobulin 
Fractions to SVT2 Cellular Preparations 







Binding {%Y 








SVT2ccUs 
in culture 


SVT2 tumor 
pieces from immuno- 
suppressed mice 


SVT2 tumor pieces 
from nonimmuno- 
suppressed mice 


»^'I-Normal y-globuUn 

'"I-lgGu 
«»I-Normal y-globulin 

^MgG« 
»»I.Normal IgG,b 


29.5 
25.8 
15.8 


26.9 

3.2 

21.8 
2.9 

22.9 
4.8 




11.7 
1.9 

11.3 
3.3 

5.5 
3.8 



"1x10^ SVT2 cells maintained in tissue culture or 0. 1 g of a crude cellular preparation made by finely mincing 
tumors taken from immunosuppressed (400 R X-ray and cortisone) or nonimmunosuppressed mice, were in- 
cubated with a minimum of 2000 cpm of radiolabeled antibody or normal y-globulin, and the cpm bound/cpm 
added initially (average of dupUcate samples) determined. 



62 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



from SV40 tumor cells in vivo. These 
antigens reappeared after the cells were 
cultured in vitro. In contrast, our results 
show that '^Mabeled tumor-specific anti- 
bodies can be prepared from the sera of 
mice with progressively growing SVT2 
tumors, and that when passively trans- 
ferred these antibodies localize specifically 
in SVT2 tumors growing in immunosup- 
pressed (X-ray and cortisone) mice. This 
localization was blocked by passive ad- 
ministration of anti-SVT2 serum or by 
nonimmunosuppression of the host. We 
used the F, hosts and SVT2 tumor to 
exactly duplicate the model and results re- 
ported by Ting (10, II). The specificity of 
the antisera obtained from Fi hosts was 
tested by absorption experiments. Their re- 
activity could only be abrogated by absorp- 
tion with various SV40-transformed cells 
(10, II). To assure that we were dealing 
with a tumor-specific antigen and not a 
histocompatibility antigen, we obtained the 
SVT2 tumor frozen from a tumor bank. In 
addition, the Fi mice were from the same 
source at the NCI, where a new colony is 
started each year to prevent genetic drift. 
Finally, the antibody did not show any 
cross-reactivity with other tumors of 
C57B1/6 or Balb/c background (Table I). 

The results of the present study suggest 
that the presence of antibody, produced for 
example by an intact immune system, can 
block the localization of radiolabeled 
syngeneic tumor-specific antibodies to the 
SV40 tumor-specific cell surface antigen in 
vivo. The findings that immunosuppressed 
mice given anti-SVT2 serum had significant 
free antibody in the circulation and that the 
radiolabeled antibody fractions showed 
lower binding to tumors taken from non- 
immunosuppressed mice, suggest that de- 
position of unlabeled antibody in tumors 
masks or modulates (21) antigenic sites on 
the SV40 tumor cell surface and prevents 
their recognition by radiolabeled tumor lo- 
calizing antibodies. Witzr/ al. (22) reported 
that irradiation (400 rad) of the host prior to 
MDAY mcthylcholanthrene-induced sar- 
coma cell inoculation caused a decreased 
coating of tumor cells with immunoglobu- 
lin. Ran {'/ (//. (23) elutcd cytotoxic anti- 



tumor antibodies from SEYF-a tumor ceUs 
using low pH buffers. However, acid ela- 
tion gave no direct evidence of the presence 
of antibody on the SVT2 tumor in situ. Izzo 
et al. (24) reported that horse anti-rat lym- 
phocyte serum blocked the localization in 
allogeneic skin grafts of radiolabeled his- 
tocompatibility antibody. 

In Table IV, there was an increase of 
'^'I-labeled normal gamma globulin lo- 
calization in SVT2 tumor in the im- 
munosuppressed and immunosuppressed + 
antiserum groups relative to the nonim- 
munosuppressed group. This can probably 
be explained by the well-known increase in 
skin vascular permeability following X-ray 
(25). The localization of normal gamma 
globulin in the SVT2 tumor was not re- 
duced in the immunosuppressed mice given 
antiserum (Group III) compared with im- 
munosuppressed mice (Group II), but was 
reduced somewhat in muscle and skin. The 
reason for this reduction is not clear. How- 
ever, it should be kept in mind that these 
differences involve tissue to blood ratios 
that are quite uniform and in most instances 
considerably less than 1, whereas the SVT2 
tumor to blood ratio is greater than 2 at 48 
hr in Group II. Ballou et al. (1, 26) reported 
that radiolabeled monoclonal antibody lo- 
calized preferentially in a mouse teratino- 
carcinoma in syngeneic mice, with a tumor 
to blood ratio of 1.3 at 48 hr after in- 
jection. This ratio increased to IS at S 
days after injection due to the rapid elimi- 
nation of the IgM antibody from the blood. 

Present studies using the SVT2 model 
have limitations in replicating what is pres- 
ently being attempted clinically. First, to 
date, most radiolabeled antisera have been 
of heterologous derivation, goat and rabbit 
into man rather than comparable to the ex- 
perimental situation of murine antibody 
into a murine host. In addition, most of the 
antigens studied in man have been tumor 
associated, i.e., in high concentration but 
not unique, nor immunogenic. HCG,^ CEA, 
ferritin and AFP are generally not consid- 

^ Abbreviations used: HCG, human chorionic go- 
nadotropin; CEA, carcinoembryonic antigen; AFP, 
«-fetoprotein. 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



63 



ered immunogenic to the host nor have they 
been clearly demonstrated to produce host 
autoantibody. 

The results suggest that a possible limita- 
tion in the use of radiolabeled antibodies for 
cancer diagnosis and therapy might be the 
production by the host of antibodies that 
bind to antigen sites on the tumor cell sur- 
face and block binding of labeled antibody. 
This would only be relevant to those anti- 
gens which induce autoantibody and only 
then if the heterologous-derived antibody 
would be similarly blocked. However, 
there is a high probability that monoclonal 
antibodies produced by hybridomas of hu- 
man origin will eventually be used in cancer 
diagnosis and therapy, so that the results in 
mouse systems will be of importance in 
indicating the directions human studies 
should take. 

We acknowledge Dr. William F. Bale (deceased, 
June 28, 1982) for helpful advice and suggestions. 
Dr. Eugene A. Johnson for help with the statistical 
analysis. Miss Lezlie A. Nelson for technical assis- 
tance, and Mrs. Peggy Evans for typing of the manu- 
script. This work was supported by Grants CA- 
23967 and CA- 15548 from the National Cancer Insti- 
tute. 

1. Ballou B, Levine G, Hakala TR, Solter D. Tumor 
location detected with radioactively labeled 
monoclonal antibody and external scintigraphy. 
Science 206:844-847, 1979. 

2. Buchsbaum DJ, Walker PC, Johnson EA. Dis- 
tribution of radiolabeled alloantibodies in mice 
bearing 3-methylcholanthrene-induced sarcomas. 
Cancer Res 39:3363-3368, 1979. 

3. Ghose T, Norvell ST, Aquino J, Belitsky P, Tai J, 
Guclu A, Blair AH. Localization of >^4-labeled 
antibodies in human renal cell carcinomas and in a 
mouse hepatoma and correlation with tumor de- 
tection by photoscanning. Cancer Res 40:3018- 
3031, 1980. 

4. Goldenberg DM, Kim EE, DeLand FH, Bennett 
S, Primus FJ. Radioinmiunodetection of cancer 
with radioactive antibodies to carcinoembryonic 
antigen. Cancer Res 40:2984-2992, 1980. 

5. Mach J-P, Buchegger F, Fomi M, Ritschard J, 
Berche C, Lumbroso J-D, Schreyer M, Girardet 
C, Accolla RS, Carrel S. Use of radiolabelled 
monoclonal anti-CEA antibodies for the detection 
of human carcinomas by external photoscanning 
and tomoscintigraphy. Immunol Today 2:239- 
249, 1981. 



6. Order SE, Klein JL, Ettinger D, Alderson P, 
Siegelman S, Leichner P. Use of isotopic immu- 
noglobulin in therapy. Cancer Res 40:3001-3007, 
1980. 

7. Pressman D. The development and use of 
radiolabeled antitumor antibodies. Cancer Res 
40:2960-2964, 1980. 

8. Ting C-C, Herberman RB. Detection of tumor- 
specific cell surface antigen of Simian virus 40- 
induced tumors by the isotopic antiglobulin tech- 
nique. Int J Cancer 7:499-506, 1971. 

9. Ting C-C, Rodrigues D, Herberman RB. Expres- 
sion of fetal antigens and tumor-specific antigens 
in SV40-transformed cells. IL Tumor transplanta- 
tion studies. Int J Cancer 12:519-523, 1973. 

10. Ting C-C, Ortaldo JR, Herberman RB. Serologic 
analysis of the antigenic specificities of Simian 
virus 40- transformed cells and their relationship to 
tumor-associated transplantation antigen. J Nat 
Cancer Inst 52:815-821, 1974. 

11. Ting C-C. Detection of anti-tumor antibody in vi- 
rally induced tumors and its relationship to tumor 
growth. Int J Cancer 18:205-215, 1976. 

12. Primus FJ, Bennett SJ, Kim EE, DeLand FH, 
Zahn MC, Goldenberg DM. Circulating immune 
complexes in cancer patients receiving goat 
radio localizing antibodies to carcinoembryonic 
antigen. Cancer Res 40:497-501, 1980. 

13. Izzo MJ, Bale WF. Preferential localization and 
rate of loss of labeled alloantibody from rat 
tumors and skin transplants carrying the corre- 
sponding alloantigen. Cancer Res 36:2868-2873, 
1976. 

14. New and views. Nature (London) 248:637, 1974. 

15. Sparks FC, Ting CC, Hammond WG, Herberman 
RB. An isotopic antiglobulin technique for 
measuring antibodies to cell-surface antigens. J 
Immunol 102:842-847, 1969. 

16. Bale WF, Helmkamp RW, Davis TP, Izzo MJ, 
Goodland RL, Contreras MA, Spar IL. High spe- 
cific activity labeling of protein with T** by the 
iodine monochloride method. Proc Soc Exp Biol 
Med 122:407-414, 1966. 

17. Ehriich R, Witz IP. The elution of antibodies from 
viable murine tumor cells. J Immunol Methods 
26:345-353, 1979. 

18. Bale WF, Contreras MA, Izzo MJ, Delia Penta D, 
Buchsbaum DJ. Preferential in vivo localization of 
*^I-labeled antibody in a carcinogen-induced 
syngeneic rat tumor. Prog Exp Tumor Res 
19:270-283, 1974. 

19. Ey PL, Prowse SJ, Jenkin CR. Isolation of pure 
IgGi, IgG2a and IgGsb immunoglobulins from 
mouse serum using protein A-sepharose. Im- 
munochemistry 15:429-436, 1978. 

20. Witz I, Klein G, Pressman D. Moloney lymphoma 
antibodies from mice; localization in spleens of 



64 



ANTIBODY LOCALIZATION IN SVT2 TUMOR 



Moloney lymphonui bearing mice. Proc Soc Exp 
Biol Med 130:1102-1105, 1969. 

21. De Vaux Saint Cyr Ch. Modulation of expretfion of 
virufr-induced tumor antigen* m vivo and m Wrro. 
Ur J Med Sci 10:826-835, 1974. 

22. Witz IP, Kinamon S, Ran M, Klein G. Tumour- 
bound immunoglobulins. The in vitro fixation of 
radioiodine-labelled anti-immunoglobulin re- 
agent* by tumour cells. Clin Exp Immunol 
16:321-333, 1974. 

23. Ran M, Klein G, Witz IP. Tumor-bound immuno- 
globulins. Evidence for the in vivo coating of 
tumor cells by potentially cytotoxic anti-tumor 
antibodies. Int J Cancer 17:90-97, 1976. 

24. lzzoMJ,ContreraftMA, Bale WF. Localization in 



rat skin transplants of purified ■**l-ljibeled 
xenogeneic histocompatibility antibody. Pkoc Soc 
Exp Biol Med 154:184-187, 1977. 

25. Law MP. Thomlinson RH. Vascular permeability 
in the ears of rats after X-irradiation. Brit J 
Radiol 51:895-904, 1978. 

26. Levine G, Ballou B, Reiland J, Softer D, Gumer- 
man L, Hakala T. Localization of I-131-labeled 
tumor- specific monoclonal antibody in the 
tumor-bearing BALB/c mouse. J Nuclear Med 
21:570-573, 1980. 



Received February 3, 1982. P.S.E.B.M. 1982, 
Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 65-71 (1982) 



Immunocytochemical Localization of Catechol-O-Methyltransf erase 
in Rat Parotid Gland (41478) 

K. INOUE,^ C. R. CREVELING, and L. W. TICE^ 

Laboratory of Experimental Pathology, and Laboratory ofBioorganic Chemistry, National Institute 
Arthritis, Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 

Bethesda, Maryland 20205 



Abstract. Catechol-O-mcthyltransferase (COMT) (EC 2.1.1.6) was localized in rat 
parotid gland using immunocytochemical methods. Immunoreactive deposits were found in 
the cytoplasm of myoepithelial cells, striated duct cells, myoepithelial-like cells of the small 
excretory duct, and small basal cells of the large excretory duct. The results confirm that the 
predominant localization of COMT is extraneuronal. After ligation of the main excretory 
duct, a marked reduction of COMT immunoreactivity was demonstrated in the striated duct. 
Chronic postganglionic sympathectomy did not produce a diminution of the COMT im- 
munoreactivity in the parotid gland. The pattern of localization observed in the ex- 
traneuronal elements suggests that enzyme may function in extraneuronal inactivation of 
catechols in the parotid gland. 



Catechol-O- methyltransferase (COMT) 
(EC 2.1.1.6) catalyzes the transfer of a 
methyl group from 5-adenosylmethionine 
to one of the phenolic hydroxyls of a vari- 
ety of catechols (1). This O-methylation 
reaction is important in the enzymatic inac- 
tivation of circulating catecholamines (2), 
the enzymatic inactivation of 2- and 4- 
hydroxyestrogens (3), the detoxification of 
xenobiotic catechols (4), and the local inac- 
tivation of catecholamines released as 
transmitters from the terminals of both 
central and peripheral catecholamine- 
containing neurons (S, 6). 

Several biochemical studies have demon- 
strated the presence of COMT in salivary 
glands (7-9), and indirect evidence has 
suggested that COMT is localized both in 
extraneuronal and intraneuronal locations 
in the salivary gland (7, 10). 

The localization of COMT in acinar and 
intercalated duct myoepithelial cells of the 
rat parotid gland has already been reported 



* Present address: Department of Oral Anatomy, 
School of Dentistry, Okayama University, Okayama, 
700, Japan. 

* To whom correspondence and reprint requests 
should be addressed: Section on Cellular Function and 
Ultrastmcture, Laboratory of Experimental Pathol- 
ogy, NIADDK, NIH, Bethesda, Md. 20205. 



using immunocytochemical methods (11). 
In the present study we have shown, using 
the peroxidase-antiperoxidase method (12), 
that COMT is also present in striated duct 
cells, myoepithelial cells of the small ex- 
cretory ducts, and in small basal cells of the 
excretory ducts. We have also examined 
the effects of sympathectomy and ligation 
of the excretory duct on the localization of 
COMT in the rat parotid gland. 

Materials and Methods. Adult Wistar 
rats (200-400 g) of both sexes were used. 
Rats were anesthetized with Nembutal and 
perfused through the heart with 100 ml of 
periodate- lysine -paraformaldehyde (PLP) 
(13) 14 days after the surgical procedures. 
After perfusion the parotid glands were rap- 
idly removed, cut into small pieces, and 
fixed with PLP for 3 hr at 4°. Tissues were 
dehydrated through graded alcohols and 
embedded in paraffin. Sections 8 /im in 
thickness were cut, mounted on glass slides 
with egg albumin and glycerine, and dried 
overnight at 37°. For immunocytochemical 
staining paraffin sections were deparaffinized 
with xylene and brought to water through 
graded alcohols. 

Slides were washed with 0.1 M phos- 
phate-buffered saline (PBS) and then treated 
with O.OOS M unbuffered periodic acid for 
S min and incubated with a solution of 



65 
0037-9727/82/090065-07$01 .00/0 

AU nihu iescfv«d. 



66 



LOCALIZATION OF COMT IN RAT PAROTID GLAND 




LOCALIZATION OF COMT IN RAT PAROTID GLAND 



67 



NaBH4 (10 mg/ml PBS) to inhibit endo- 
geneous peroxidase activity (14). Sections 
were rinsed with two changes of PBS for 
10 min and then incubated with a 1:50 dilu- 
tion (in PBS) of normal goat serum for 30 
min at room temperature. 

Slides were incubated for 24 hr at 4"* in a 
1: 1000 dilution (in PBS) of either rabbit an- 
tiserum to rat liver COMT (11), or as a 
control, in normal rabbit serum similarly 
diluted, a 1:10 dilution (in PBS) of goat 
anti-rabbit serum (Polyscience Inc., War- 
rington, Pa.) for 30 min at room tempera- 
ture and then in a 1:10 (in PBS) dilution of 
peroxidase -antiperoxidase complex (PAP) 
(Polyscience Inc., Warrington, Pa.) for 45 
min at room temperature. All slides were 
washed for 15 min each in three changes of 
PBS between each step. Peroxidase reac- 
tion product was developed by incubation 
of sections for 5 min in 3,3'-diaminobenzi- 
dine (0.05%) in 0.05 M Tris buffer, pH 7.6, 
containing 0.01% H2O2. After a final wash 
in distilled water the sections were dehy- 
drated through graded alcohols, cleared 
with xylene, and mounted in Canada Balsam. 

Atrophy of the parotid gland was pro- 
duced by ligation of the excretory duct. The 
parotid glands were postganglionically 
sympathectomized by excision of the supe- 
rior cervical ganglion. The operations were 
performed under Nembutal anesthesia. In 
all cases, the operation was unilateral, the 
intact side serving as a control. 

Results. Immunocytochemical reaction 
of COMT of the normal parotid gland 
{Figs. 1-4). The sites of positive histologi- 
cal reaction product indicate the localiza- 
tion of interaction of the specific antiserum 
with tissue COMT and will be referred to as 
a COMT-positive reaction. COMT-positive 
product was observed in the cytoplasm of 



the myoepithelial cells of the acini and in- 
tercalated ducts, myoepithelial-like cells of 
the small excretory duct, and small basal 
cells of the large excretory duct. COMT 
was also present in the striated duct cells. 
Here, however, it appeared to be as- 
sociated with basal striations. The differ- 
ences in distribution were dramatically 
shown where transitions between striated 
and intercalated ducts were present in the 
sections (Fig. 1). 

In the intralobular ducts the density of 
the COMT-positive reaction appeared 
much greater in the striated duct nearest the 
intercalated duct. The myoepithelial cells 
containing COMT-positive product were 
located at the bases of the acinar serous 
cells and intercalated duct (Fig. 1). Control 
sections of the parotid gland were negative 
(Fig. 2). 

COMT-positive reaction was also present 
in myoepithelial-like cells located at the 
bases of small excretory ducts in the inter- 
lobular connective tissue (Fig. 3). Small 
basal cells containing COMT were found in 
the epithelium of the large excretory ducts 
(Fig. 4). 

Immunocytochemical reaction of COMT 
following the surgical procedure {Figs. 
5-9). 1. Superior cervical ganglionectomy 
{Figs. 5-7). Fourteen days after this proce- 
dure no histological changes were ob- 
served. COMT-positive product was pres- 
ent in the cytoplasm of the striated duct 
cells, myoepithelial cells (Figs. 5, 6), and 
small basal cells of the large excretory duct 
(Fig. 7). This distribution was similar to that 
of the normal parotid gland. 

2. Ligation {Fig. 8). Ligation of the main 
excretory duct caused the expected histo- 
logical changes in the parotid gland (see 
Refs. (10, 15)). The acinar cells not only 



Fio. 1. Normal rat parotid gland. Strong specific staining for COMT is seen in the cytoplasm of the 
myoepithelial cells (arrows) and striated duct cells (SD) near the intercalated duct. A, acini; ID, 
intercalated duct. xl45. 

Fig. 2. Normal rat parotid gland treated with normal rabbit serum as control. No staining for COMT 
is seen. xl45. 

Fig. 3. Normal rat parotid gland. COMT is present in the myoepithelial-like cells of the small 
excretory duct (ED). x410. 

Fio. 4. Normal rat parotid gland. Small basal cells (arrows) of the large excretory duct (ED) contain 
COMT. X655. 



fA 



\.f)f.\i.iy.\l\f)^ OF COMT IN RAT PAROTTD GLA\D 













' c 



) 1 "' 



• - . - 



LOCALIZATION OF COMT IN RAT PAROTID GLAND 



69 



atrophied, but also decreased in number. 
The intralobular and interlobular ducts had 
atrophied and their lumina were markedly 
dilated and the interstitial connective tissue 
had proliferated. The myoepithelial cells 
were intact and had COMT-positive reac- 
tion products (Fig. 8). 

i. Ligation and postganglionic sym- 
pathectomy (Fig, 9). The glands that had 
been ligated and sympathectomized were 
similar to that seen after ligation alone. 
COMT was observed only in the myoepi- 
thelial cells (Fig. 9). 

Discussion. In classical experimental 
approaches, salivary gland functions are 
differentiated by examination of the effects 
ot ligation of the major excretory ducts or 
denervation of the gland. Following ligation 
the major secretory and ductal components 
of the gland, with the exception of the 
myoepithelial cells, undergo a striking atro- 
phy (IS, 16). After surgical, chemical, or 
immunological sympathectomy those func- 
tions of the gland dependent upon sympa- 
thetic innervation are lost (17, 18). 

Previous studies have shown that rodent 
salivary glands contain both catechol- 
amines and the enzymes, monoamine 
oxidase and COMT, responsible for their 
metabolic inactivation (17-20). Following 
sympathectomy, factors specifically related 
to intact sympathetic innervation such as 
the intraneuronal biosynthetic enzymes, 
tyrosine hydroxylase, and dopamine-j3- 
hydroxylase, the tissue level of norepineph- 
rine, and the high-affinity uptake system for 
norepinephrine all decline in parallel (18, 
20). After atrophy due to duct ligation these 
factors remain essentially unchanged after 
appropriate adjustments are made for the 
reduction in glandular mass (10, 20). 



Sympathectomy results in minimal re- 
ductions in monoamine oxidase levels while 
duct ligation leads to a profound decrease in 
enzyme levels. While it has been clearly 
established that monoamine oxidase is 
present in sympathetic neurons it is also 
widely distributed in other cell types in the 
salivary gland. Thus the minimal reduction 
in activity following degeneration of sym- 
pathetic neurons is consistent with the dis- 
tribution of the enzyme. In the case of 
COMT, sympathectomy results in very 
minor changes in enzyme activity in the 
parotid gland (IS). Similar studies in the 
submaxillary gland of rat indicate a sym- 
pathectomy-dependent decrease in COMT 
activity and suggest the presence of intra- 
neuronal COMT (8). 

However, duct ligation results in a partial 
and quite variable reduction of COMT ac- 
tivity ranging from 10 to 40% of the activity 
in unligated glands. A major fraction of 
COMT activity is unaffected in the pres- 
ence of a major atrophy of the gland (8, 
IS, 20). 

Our present results provide direct, visual 
confirmation of these conclusions. As in 
other tissues studied (11, 21, 22), we have 
found no evidence for a neuronal localiza- 
tion for COMT. In the salivary gland, 
COMT was found in the striated duct cells, 
the small basal cells of the excretory ducts, 
and in myoepithelial cells. In those loca- 
tions the enzyme is not detectably affected 
by sympathectomy. Ductal ligation led to a 
variable loss in COMT located in striated 
duct cells and no detectable change in 
COMT in myoepithelial cells. The myoep- 
ithelial localization of COMT was un- 
changed following both duct ligation and 
sympathectomy. Thus the variable de- 



FiG. 5. Rat parotid gland after removal of the superior cervical ganglion. COMT reaction products 
are seen in the cytoplasm of the myoepithelial cells (arrows) and striated duct cells (SD). A, acini; IR, 
intercalated duct, x 190. 

Fig. 6. Rat parotid gland after removal of the superior cervical ganglion. Specific staining for 
COMT is observed in the myoepithelial-like cells of the small excretory duct (ED). x425. 

Fig. 7. Rat parotid gland after removal of the superior cervical ganglion. COMT reaction product 
are found in the small basal cells of the large excretory duct (ED). x535. 

Fig. 8. Rat parotid gland following ligation of the main excretory duct. The myoepithelial cells 
(arrows) contain COMT. A, acini. x410. 

Fig. 9. Rat parotid gland following ligation of the excretory duct and removal of the superior 
cervical ganglion. The myoepithelial cells (arrows) contain COMT. A, acini. x400. 



70 



LOCALIZATION OF COMT IN RAT PAROTID GLAND 



crease in COMT activity following duct li- 
gations may reflect the loss of striated duct 
cells and possibly the small basal cells of 
the excretory duct. The major fraction of 
COMT activity which is unaffected by 
ligation-induced activity clearly appears to 
be associated with myoepithelial cells. 

The precise function of COMT in these 
extraneuronal sites is problematic. There is 
little information about the functional role 
of basal cells in the salivary gland excretory 
ducts. A similar question can be raised con- 
cerning the role of COMT in the small, 
basal cells of the rat epididymis (11). With 
regard to myoepithelial cells on the other 
hand, conventional thinking suggests that 
the presence of COMT in contractile 
myoepithelial cells is related to the inacti- 
vation of norepinephrine released locally 
from noradrenergic varicosities or perhaps 
by direct sympathetic innervation of 
myoepithelial cells. Thus, by analogy with 
cholinergic nerve -muscle relationships, 
COMT has been presumed to be the norad- 
renergic analog of acetylcholinesterase. 

Certain aspects of this analogy are of 
interest. The contractile function of myo- 
epithelial cells, first suggested as early as 
1661 (23), is well documented (24). These 
cells contain filaments similar to those of 
smooth muscle cells; they respond to sym- 
pathetic stimulation by contraction leading 
to increased ductal pressure. The response 
is mediated through an a-adrenergic re- 
ceptor and has been observed directly in 
organ culture. Further, it is known that the 
acini of rat parotid glands are closely as- 
sociated with catecholaminergic, fluores- 
cent varicosities. It is of interest to note 
that a fraction of norepinephrine uptake by 
salivary gland is reserpine resistant, sug- 
gesting a significant participation of ex- 
traneuronal uptake sites for norepineph- 
rine. Furthermore, reserpine treatment re- 
sults in a partial loss of COMT activity 
within 6- 18 hr (18). These findings suggest 
that both norepinephrine uptake and the 
reserpine-induced reduction of COMT ac- 
tivity may occur in myoepithelial cells. The 
possibility that the reserpine-sensitive frac- 
tion of COMT in salivary glands is present 



in myoepithelial cells is under examination 
by our immunocytochemical procedure to 
determine whether this effect occurs in 
myoepithelial cells or elsewhere. 

Recent experimental results have greatly 
weakened any direct functional or mor- 
phological analogy between acetylcholin- 
esterase and COMT. Localization of 
COMT in brain by fluorescent immunocy- 
tochemical techniques have demonstrated 
glial and ependymal cell localization rather 
than a neuronsd localization. Other nuyor 
sites of COMT are the choroid plexus (22) 
and the ciliary body of the eye (25). It has 
been suggested that in these sites COMT 
functions as a barrier or ''enzymatic sink" 
to the passage of free catechols which re- 
strict neuronally released catechols to a 
local area (25). Studies of the localization of 
COMT in the reproductive tract suggest a 
similar function of the enzyme as a protec- 
tive barrier or '*enzyme-sink" for catechol- 
estrogens (26). Finally it should be noted 
that COMT has been demonstrated inmiuno- 
cytochemically in the ductal cells of the 
lactating rat mammary gland where it may 
play the same role as in the salivary gland 
(27, 28). Thus it is clear that further investi- 
gations will be required to define the nature 
of the role of COMT in salivary gland myo- 
epithelial cells with certainty. 



We wish to thank Mrs. Cahill, Mr. Carter, and Mr. 
Nakamura for their skilled technical assistance, and 
Ms. Colleen LePore for her clerical assistance. 

1. Axelrod J. Methylation reactions in the formation 
and metabolism of catecholamines and other 
biogenic amines. Pharmacol Rev 18:95-113, 
1966. 

2. Kopin J, Axelrod J, Gordon E. The metabolic fate 
of H^-epinephrine and C"-metanephrine in the 
rat. J Biol Chem 236:2109-2113, I%1. 

3. Merriam GR, MacLusky NJ, Johnson LA, Nafto- 
lin F. 2-Hydroxyestradiol-17af and 4-hydroxyes- 
tradiol-lTor, catecholestrogen analogs with re- 
duced estrogen receptor affmity. Steroids 36:13- 
20, 1980. 

4. Creveling CR, Dalgard N, Shimizu H, Daly JW. 
Catechol-0-methyltransferase 111: Meta- and 
para-O-methylation of catechols. Mol Pharmacol 
6:691-696, 1970. 



LOCALIZATION OF COMT IN RAT PAROTID GLAND 



71 



5. Guldberg HC, Marsden CA. Catechol-O- 
methyltransferase: Pharmacological aspects and 
physiological role. Phannacol Rev 27:135-206, 
1975. 

6. Jarrott B. Occurrence and properties of 
catechoI-O-methyltransferase in adrenergic 
neurons. J Neurochem 18:17-27, 1971. 

7. Iversen LL, Glowinski J, Axelrod J. The 
physiologic disposition and metabolism of norepi- 
nephrine in immunosympathectomized animals. J 
Pharmacol Exp Ther 151:273-284, 1966. 

8. Broch OJ. Effect of reserpine on catechol-O- 
methyltransferase in rat submaxillary gland. J 
Pharm Pharmacol 26:375-377. 1974. 

9. Ikeno T, Ikeno K, Hashimoto S, Nemoto K, 
Hasegawa J, Kuzuya H. Catecholamine metabo- 
lism in salivary glands of the rats. 3. Devel- 
opmental changes in catechol-O-methyltransfer- 
ase activity. J Tohoku Dent Univ 4:82-85, 1977. 

10. Anden N-E, Norberg K-A, Ison L. The adrenergic 
nerves of rat rat salivary glands after excretory 
duct ligation. Acta Physiol Scand 66:501-506, 
1966. 

11. Inoue K, Tice LW, Creveling CR. Immunocyto- 
chemical localization of catechol-O-methyltrans- 
ferase. In: Usdin E, Weiner N, Youdim MBH, 
eds. Biochemistry and Function of Monoamine 
Enzymes. New York, Dekker, pp835-859, 1977. 

12. Stemberger LA. Enzyme immunocytochemistry. 
In: Hayat MA, ed. Electron Microscopy of En- 
zymes. Principles and Methods. New York, Van 
Nostrand Reinhold, ppl50-191, 1973. 

13. McLean IW, Nakane PK. Periodate- lysine - 
paraformaldehyde fixative, a new fixative for im- 
munoelectron microscopy. J Histochem Cyto- 
chem 22:1077-1083, 1974. 

14. Isobe Y, Chen ST, Nakane PK, Brown WR. 
Studies on translocation of immunoglobulins 
across intestinal epithelium I. Improvements in 
the peroxidase-labeled antibody method for appli- 
cation to study of human intestinal mucosa. Acta 
Histochem Cytochem 10:161-171, 1977. 

15. Marsden CA, Broch OJ, Guldberg HC. Catechol- 
O-methyltransferase and monoamine oxidase ac- 
tivities in rat submaxillary gland: Effects of liga- 
tion, sympathectomy and some drugs. Eur J 
Pharmacol 15:335-342, 1971. 

16. Garrett JR, Kemplay SK. The adrenergic inner- 
vation of the submandibular gland of the cat and 
the effects of various surgical denervations on 
these nerves. A histochemical and ultrastructural 
study including the use of 5-hydroxydopamine. J 
Anat 124:99-115, 1977. 

17. Haggendal J. The disappearance of dopamine-/3- 
hydroxylase from rat salivary gland after extirpa- 



tion of the superior cervical ganglion. J Neural 
Transmission 48:249-259, 1980. 

18. Kuzuya H, Ikeno T, Ikeno K, Memoto K, 
Hashimoto S. Catecholamine contents and ac- 
tivities of catecholamine synthesizing and inac- 
tivating enzymes in the salivary glands of young 
growing rats. Arch Oral Biol 25:31-36, 1980. 

19. Almgren O, Anden N-E, Jonason J, Norberg K-A, 
Olson L. Cellular localization of monoamine 
oxidase in rat salivary glands. Acta Physiol Scand 
67:21-26, 1966. 

20. Jonason J. Metabolism of dopamine and nor- 
adrenaline in normal, atrophied and post- 
ganglionically sympathectomized rat salivary 
glands in vitro. Acta Physiol Scand 76:299-311, 
1%9. 

21. Kaplan GP, Hartman BK, Creveling CR. Im- 
munohistochemical demonstration of catechol- 
O-methyltransferase in mammalian brain. Brain 
Res 167:241-250, 1979. 

22. Kaplan GP, Hartman BK, Creveling CR. Anti- 
catechol-O-methyltransferase: Demonstration of 
specificity and immunological cross reactivity 
with the enzyme from rat liver, kidney, brain and 
choroid plexuses. Neurochem Res 5:869-877, 
1980. 

23. Stenonis N. Glandulis oris et novis earundem 
vasis, 1661. Bodlei Library, Oxford. 

24. Garrett JR, Emmelin N. Activities of salivary 
myoepithelial cells: A review. Med Biol 57:1-28, 
1979. 

25. Creveling CR, Hartman BK. Relationships be- 
tween the cellular localization and the physiologi- 
cal function of catechol-O-methyltransferase. In: 
Borchardt RT, Creveling CR, Usdin E, eds. Pro- 
ceedings of the Conference on Transmethylation, 
Lake of the Ozarks, October 26-29, 1981. Lon- 
don, Macmillan, pp479-486, 1982. 

26. Inoue K, Tice LW, Creveling CR. Immunocy to- 
chemical localization of catechol-O-methyltrans- 
ferase in the pregnant rat uterus. Endocrinology 
107:1833-1840, 1980. 

27. Lowe MC, Amin AF, Creveling CR. Immunocy- 
tochemical localization of catechol-O-methyltrans- 
ferase in normal and cancerous breast tissues. In: 
Borchardt RT, Creveling CR, Usdin E, eds. Pro- 
ceedings of the Conference on Transmethylation, 
Lake of the Ozarks, October 26-29, 1981. London, 
MacmiUan, pp487-490, 1982. 

28. Strum JM. Estrogen-induced alterations in the 
myoepithelial cells of the rat mammary gland. Cell 
Tissue Res 193:155-161, 1978. 

Received November 30, 1981. P.S.E.B.M. 1982, 
Vol. 171. 



IWXJbl.tMNO^ Of rut WICI£T V FOIt IXftMIMtfilAL BKHXlCn <^VD MEJMCHtX 171. T2-7B iHC) 



Light and Electron Microscopic Studies of the Paltiogenesis of Vs 

Infection in Mouse Brain (41479) 



Virus 



DIANE C. BOSSE. WALLACE G. CAMPBELL. Je-. and WILUAM A, CASSEL> 

iJeiMiftmens of MunMMoiog) and immufuAc^y . and Oepartmtni of Faikoiag} tmd Ijtkarwaary 
hmury Cnirerut} Sihooi of Uedkine. Ajlama, GeorfiaS0322 



Ah^trat t . VaicciiKui viruh rcpbcuokm in i^amHa^ woutt bnan% m-as ^»»— i^*^ b% 1 
electroo fiucfo«cof>y 2 tu 5 dai>% after iooculalioiL ix.. dunqg the pcricMl of i 
iotectivjty. KepftiocLioo wteb were detected in nifjiingfjil celk. ad^emitii] ceils of i 
artefkfleb. and wnadl nooneuroiia] cefli of the bnin. So evidence ma& found for lepftcaiion in 
neuroob. altfaougli tbe%r cell% »ere altered indvectl). The virus can be described 
accurately ab being J 



The fuituraJ hittory of poxvirus infection 
UMiaJly includes involvement of the respi- 
niory tract, reticuloendothelial system, 
liver, and skin. Infection of the brain is 
rare, but is of interest because of the possi- 
bility that encephalitis following immuniza- 
tion with vaccinia virus may be a manifes- 
tation of infection. It is unlikely that it re- 
sults from a toxic effect i\). Experimental 
inoculation of brain with vaccinia, virus has 
revealed that some strains proliferate in 
nervous tissue. Uncertainties exist, how- 
ever, as to the types of cells involved and 
the natural hisu>ry of intracerebral infec- 
tion. Several reports have appeared sug- 
gesting that the principal lesions of cerebral 
infections are a leptomeningitis and/or 
ependymitis (2- 5). Most of these studies 
have been conducted in mice, and the ap- 
proach has been principally through light 
microscopy, including the application of 
imniunofluorescence. An exception to the 
usual reports has been a study in which a 
raci)on poxvirus ini>culated into mice was 
fi)und to cause necrotic foci in dorsal spinal 
ganglia and severe demyclination with ne- 
crosis in the lumbar plexuses (6). These 
fmdings suggest a modification of neurons 
due to a direct viral infection. Where it has 
been considered, there is no evidence for 
viral replication in glial cells. Immunofluo- 
rescence microscopy is not sensitive 
en4>iigh to justify the conclusion that 
iininunolluorescent-negative cells are not 

' Id wluim icpiiul icqiicsls shoulil be addressed. 



infected. Conversely, positive immimofluo- 
rescence does not guarantee intracellular 
viral replication. Electron microscopy per- 
mits direct observation of the characteristic 
morphology of poxviruses in both their 
replicating and mature forms. Accordingly, 
electron microscopy was brought to bear on 
the question of poxvirus neurovirulence 
and focused on the early period of maximal 
infectivity after the intracerebral inocula- 
tion of mice with vaccinia virus. 

Materials and Methods. Poxvirus. Vac- 
cinia virus, strain Levaditi-S8 (7), is a virus 
capable of multiplying to a high level in 
mouse brain. It was assayed by dropping 
the virus on the chorioallantois of 11-day 
old chicken embryos, and after 2 days of 
incubation at 37° the results were scored as 
pock-producing units (ppu) per 0.05 ml of 
inoculum. 

Mice. Three-week-old, NIH general pur- 
pose albino mice were employed. Virus was 
administered by injecting 0.02 ml of sus- 
pension into the right cerebral hemisphere 
of ether-anesthetized mice. 

Viral replication in brain. Eighteen mice 
were inoculated with a selected virus dose 
and three mice were sacrificed each day 
thereafter, at which time the pooled brains 
were ground with RR Alundum and brought 
to a 20% suspension, by weight, in 0.85% 
NaCl solution. From the growth curve re- 
sults, a suitable virus dose (100 ppu) was 
selected for further studies. Control mice, 
sham inoculated with NaCl solution, were 
run in parallel. 



72 

'1 *)7:7 k: (HXH)7: o7M)i.(H)/o 

|hl ,( I IVM.* hy ihr SiKieiy Un I t^pcmnenlal Hiology Mnd Medicine, 
ft /#»♦<•/ • «•«/ 



VACCINIA VIRUS PATHOGENESIS 



73 



Preliminary preparation of infected 
brain. For 6 days after intracerebral inocu- 
lation with 100 ppu of virus, each of two 
mice was anesthetized, and an inftision was 
begun through the left ventricle of the heart 
with 20 ml of cold (4**) imidazole -formalin 
solution (0.1 M imidazole- HCI, 10% for- 
malin, pH 7.2). As the infusion was begun, 
the hepatic vein was cut. After infusion, the 
brain and spinal cord were dissected out 
and stored in imidazole -formalin solution 
at 4"* for at least 24 hr. Control mice were 
processed in an identical manner. 

Fixation and sectioning of brain. The 
cold, fixed brain was cut coronally into 
eight serial sections. Subsections for light 
microscopy were embedded in Paraplast 
and the caudal surface was sectioned to a 
thickness of 5 fim. The sections were 
stained with hematoxylin and eosin. 

The specimens for electron microscopy 
were divided ftirther by a medial cut, di- 
viding them into right and left sides of the 
brain. Prior to postfixation, ftirther subsec- 
tions were made. The subsections were 
diced into pieces less than 1 mm in greatest 
diameter and were ftxed in isoosmolal, 
buffered, 1.5% glutaraldehyde, pH 7.4, for 
3 hr at 4"" and then placed in a wash solution 
(isoosmolal sucrose solution with 0.02 M 
imidazole- HCI, pH 7.4) overnight at 4°. 
(The tissue used for electron microscopy 
had been kept at 4"* until this point.) After a 
1-hr exposure to 2% OSO4 in sodium barbi- 
tal buffer at pH 7.4, the specimens were 
dehydrated through an alcohol series, car- 
ried in propylene for two 15-min periods, 
held for 30 min in a 50/50 propylene - 
Maraglas mixture and embedded in Mara- 
glas as described previously (1). Thin sec- 
tions (about 500 nm) were cut with glass 
knives, and areas to be studied by electron 
microscopy were selected. Ultrathin sec- 
tions (about 75 nm) were cut with diamond 
knives and placed on uncoated copper 
grids, stained with lead citrate, and studied 
by electron microscopy (1). 

Results. Viral replication in brain. Vac- 
cinia virus replication at three different 
doses is shown in Fig. 1. The virus repli- 
cates appreciably by 3 days in mouse brain. 
Accordingly, for electron microscopic 



Vaccinia 


Virus 


(Levaditi 


-58) 






if / 


/ ' 




""^^^ / 






5000 Dpu 

500 ppu 

100 ppu 



107 

- 10® 

9 105 
o 

:io3 

o 

"102 

101 



Days Postinoculation 
Fig. 1. Replication of vaccinia vims in mouse brain 
following intracerebral inoculation with three different 
doses of virus. 

Studies the combination of a dose of 100 
ppu and an examination period of 2-4 days 
postinoculation was selected as conditions 
under which viral replication would be ex- 
pected to be extensive and yet tissue 
breakdown would be at a minimum. 

Signs of disease. At the 100 ppu dose, 
signs of disease began to appear at 3 days 
postinoculation. The mice appeared lethar- 
gic, hunched and trembling. Flaccid paraly- 
sis was noted, and body spasms occurred 
when the mice were twirled by their tails. 
By the fourth day, trembling continued and 
breathing became labored. The front legs 
no longer were used for locomotion. 
Animals surviving to the fifth day became 
totally immobilized, very sensitive to noise 
vibrations, and usually died in spasms. At 
this time, the brains were of a very soft 
consistency and covered with petechial 
hemorrhages. Some mice survived to Day 
6, but none survived longer. 

Light microscopy of infected brains. 
Histologically, the brain was essentially un- 
remarkable during the first 2 days post- 
inoculation (Fig. 2A). By the third and 
fourth days, many cortical areas showed 
nuclear shrinkage associated with early, 
occasionally marked, perinuclear clearing, 
especially in the neuronal layers (Fig. 2B). 
More extensive perinuclear clearing was 
present by the fifth day. A focal lep- 
tomeningitis and ependymitis, both of 
which included a polymorphonuclear 







m^ ^ 4i ^'•^"••i 



• 






ML 



A 





'• « 



.k 








n 







A «. 



. >^ 



• » 



* 
'i * 



INL -^ 



-♦•** 



V 



^ ^ 
• - 



,» * .\ 





1 


■• ;t 


ft 


•• 


,<'»> 




4 


Vvt* ^ 




ML 


^w; 


^1 


■ * k 


'K r* 




^ m 





— »— ^f^^^^ll II ■! ■!■■ MW 1^7..* f 

Fig. 2* Light microscopy of dorsal suiftu^e of parielal con iniu lXj One i.ki> Lifier vims 

inoculalion. Meninges (ML molecular bycr(MLK and inner neuronal layer (INL) appear normal. (B) 
Four days post inocu tat ion. Nuclear shrinkage* nuclear hyperchromasia* and periiiuclear clearing are 
evident tn cells of the molecular and inner neuronal layers, especially cells that api^ear to be neurons* 
The neuropil has a spongy quality. Note that the meninges appear unremarkable. <C) Six days post- 
inoculation. Changes in molecular and inner neuronal layers persist. In addition, an inflammatory 
infiltrate, rich in neutrophils, involves the meninges. Hematoxylin and eosin, ^clSO. 



74 



VACCINIA VIRUS PATHOGENESIS 



75 







I 




V" 



MC 



:l.'^'' 



T$, 




n» %. 



^5>- 



CP 



««Qh. 




Hi y 

Fig. 3. Three days after virus inoculation. Meninges and sybnieningeul twrtex. Above ihe pvA (P), 
mosi meningeal cells fM) contain clusters of mature virus (V) and/or sites of viral replicaiion fr). 
Arrowheads indicate individual mature viruses within Ihe pia. Below the pia, the molecular layer neuropil 
contains swollen cytoplasmic processes (CP), one of which contains a site of viral replication tR). A 
mononuclear cell (MC) shows cytoplasmic swelling. Lead citrate. x6600. 



leukocyte infiltrate, developed between the 
fourth and sixth days. Generally, the focal 
leptomeningitis first appeared on the ven- 
tral surface and later (at 5-6 days) on the 
dorsal surface (Fig. 2C), Diffuse lep- 
tomeningitis and focal ependymitis found in 
terminal mice were occasionally associated 
with focal necrosis of the underlying brain. 

Ele c tro n m ic ro scopy of infe cted bra in s . 
Brain sections from sham-inoculated con* 
tro! mice were well preserved and unre- 
markable in appearance. 

In inoculated animals, virus was not 
readily demonstrable at 2 days. Mature 
virus and sites of viral replication were 
readily found at 3-4 days, at which time 
considerable degenerative alterations were 
noted in meningeal, neuronal^ glial, epen- 
dymal^ and vascular adventitial cells. In 



contrast, endothelial and smooth muscle 
cells of the cerebral vasculature were pre- 
served. Viral structures were most easily 
found at 3 days, which corresponds with 
the peak of infectious virus titer (Fig. I). 

Between 3 and 4 days, both mature virus 
and sites of viral replication were clearly 
identified in many cells: meningeal cells 
were extensively involved (Figs, 3, 4) and 
adventitial cells of meningeal arterioles 
were also commonly involved (Fig, 4), In 
the cerebral cortex and cerebellum, numer- 
ous small, mononuclear cells of the 
neuropil of the molecular and neuronal 
layers contained sites of viral replication 
and mature virus (Fig, 5). The virus- 
containing cells frequently were adjacent to 
capillaries and neurons. These cells were 
interpreted as probably t>eing glial cells. 









r >, 



>-/ 



*<«.■. 







Ficj. 4. Three days after virus inoculation. Section of meninges (abi>ve) and subjacent molecular 
layer. An arteriolar wall shows no virus in the capillary endothelial cells (En) or smooth muscle cells 
(SM). An adventitial cell (AC) contains a site of viral replication (R). Mature virus (arrow ) can be seen 
in the cytoplasm of an adjacent meningeal cell (M). Pia (P). Lead citrate. x64(X). 




I'M. > 

svM'liinj.' 
((■;ip). I c 



Ihii'c iI;t\N .liter \iius inociilalion. C\.Tcbi;tl Li>Hc\. A nuitinv virus panicle ishort arrow) 
»1 vinl ivplicalimi iR.i arc seen in \\\o small paiacapillarN cells showing! marked cytoplasmic 
" with Llense evli>plasm e»)ntains a mature vims particle donj: am>w). Capillary lumen 



A 

'ad eiti.ite 



• (.:(H). 



7f. 



VACCINIA VIRUS PATHOGENESIS 



77 



To determine whether neurons contained 
viral material, a number of sections of cere- 
bral cortex rich in neurons, as determined 
by light microscopy, were carefully se- 
lected and studied by electron micros- 
copy. In no instance were viral structures 
found in neuronal cells, which frequently 
showed cytoplasmic clearing and other de- 
generative changes. In addition, viral 
structures never were identified in epen- 
dymal cells, endothelial cells, smooth mus- 
cle cells of blood vessels, or cells of the 
inflammatory infiltrates. 

Discussion. Previous investigators have 
shown the presence of viral antigen in lep- 
tomeningeaJ cells of vaccinia virus-infected 
mouse brains, using immunofluorescence 
techniques (2, 4). The present study con- 
firms that the meninges are a principal site 
of intracranial replication of vaccinia virus 
during the first 4 days after inoculation of 
brain and for the first time demonstrates 
unequivocally the presence of replicating 
viral structures in meningeal cells. In addi- 
tion, virus replicates in the adventitial cells 
of blood vessels, principally meningeal ar- 
terioles, and in small nonneuronal cells of 
the cerebral cortex, presumably glial cells. 
Although ependymitis was observed, no 
virus was detected in ependymal cells, 
which frequently were focally denuded. 

Blinzinger and co-workers, using elec- 
tron microscopy, studied late stages of in- 
fection, i.e., 4-6 days postinoculation in 
adult mice (5). At that time, they found vi- 
rus only in mononuclear phagocytes and ar- 
terial adventitial cells of the meninges. No 
evidence of neuronal, glial, or ependymal 
cell infection was found. The present study 
focuses on earlier stages of infection, in 
weanling mice, in contrast to Blinzinger^s 
examination of the later stages of infection 
in adult mice. These differences in experi- 
mental design appear to account for the 
variation in findings. In fact, it is postulated 
from the results of these two studies that 
after initial infection of leptomeningeal, 
vascular adventitial, and glial cells of the 
brain, a secondary infection of late occur- 
ring inflammatory infiltrate may ensue. 

In keeping with Blinzinger et al. (S), virus 
was not found within neurons, although in 



the present study marked alterations of 
these cells were noted by both light and 
electron microscopy. The neuronal changes 
are in parallel with a report on the indirect 
action of Newcastle disease virus on the 
neurons of mice (8). Since glial cells are 
thought to influence the transmission of 
nervous activity (9), consideration of this 
point may be appropriate in the present 
study. It has been shown that in primary 
cell cultures astrocytes release factors into 
the medium that promote the growth and 
prolong the survival of rat hippocampal 
neurons (10). In this regard, in the present 
study it was noted that small nonneuronal 
cells within the cerebral cortex frequently 
contained sites of viral replication. Al- 
though marked alterations of these cells, 
probably due to viral replication, precluded 
definitive structural identification, the cells 
presumably represent glial components. 
This presumption is supported by the ab- 
sence of cellular inflammatory infiltrates, 
which might lead to the confusion of cell 
types within the brain, during the time pe- 
riod covered in this study. 

The term "neurotropic" has been applied 
to numerous viruses replicating in the 
brain. It has been suggested that this is an 
unsuitable designation for poxviruses repli- 
cating in mouse brain since they do not rep- 
licate in neurons (11). Modern dictionaries, 
however, define ''neurotropic" as desig- 
nating agents having an affinity for nervous 
tissue (12, 13), thus making the term appli- 
cable to the aforementioned poxviruses. In 
the present study, viral replication (or in- 
fection) within neurons could not be de- 
tected, even though virus replicated in non- 
neuronal cells of the brain substance. The 
virus, however, effected an indirect neu- 
rono virulence, as suggested by the ob- 
served paralysis of infected animals and the 
morphologic alterations in neurons. Care 
should be taken to recognize that some vi- 
ruses are strictly "neuronotropic," repli- 
cating only within neurons. Viruses repli- 
cating in the brain also may be described as 
"encephalitic," causing inflammation of 
the brain, as distinct from its membranes 
(14). The virus examined in this study can 
be most specifically described as being 
leptomeningoencephalitic . 



78 



VACCINIA VIRUS PATHOGENESIS 



The technical assistance of Mr. WiDy B. Thomas is 
gratefully acknowledged. 

1. Sottnek HM, CampbeD WG, Jr. Cassel WA. The 
pathogenesis of vaccinia virus toxicity. II. An 
electron microscopic study. Lab Invest 33:522- 
532. 1975. 

2. Mimms CA. Intracerebral injections and the 
growth of viruses in the mouse brain. Brit J Exp 
Pathol 41:52-59. 1960. 

3. Soekawa M. Morita C. Moriguchi R. Nakamura 
M. Neurovi rule nee of vaccinia viruses in mice. 
Zentralbl Bakteriol Parasitenk Infectionskr Hyg. 
Abt I 226:434-442. 1974. 

4. Ginsberg AH. Johnson KP. Vaccinia virus menin- 
gitis in mice after intracerebral inoculation. Infect 
Immun 13:1221-1227. 1976. 

5. Blinzinger K, Hochstein-Mintzel V. Anzil AP. 
Experimental vaccinia virus meningoencephalitis 
in adult albino mice. Virological. light micro- 
scopic and ultrastructural studies. Acta Neuro- 
palhol 40:193-205. 1977. 

6. Thomas EK. Pahner EL. Obijeski JF. Nakano JH. 
Further characterization of racoonpox virus. Arch 
Virol 49:217-227. 1975. 

7. Cassel WA. Garrett RE. Relationship between 



viral neurotropism and oncolysis. 1. Study of vac- 
cinia vims. Cancer 20:433-439, 1967. 

8. Burks JS. McFarland HF. Narayan O. A unique 
neurological disease of mice induced by a para- 
myxovirus. Newcastle disease vinis. Neurology 
25:350, 1975. 

9. Glees P. Neuroglia: Morphology and Function. 
Springfield. ID. Thomas. pp49-53. 1935. 

10. Banker GA. Trophic interactions between astro- 
glial cells and hippocampal neurons in culture. 
Science 209:809-810. 1980. 

11. Fenner F. McAuslan BR. Mimms CA. Sambrook 
J. White DO. The Biology of Animal Viruses. 
New York. Academic Press. p375. 1974. 

12. Stedmans Medical Dictionary. 23rd ed, Balti- 
more. Williams & Wilkins. 1976. 

13. Dorland*s Illustrated Medical Dictionary, 26th ed, 
Philadelphia. Saunders. 1981. 

14. Longson M. The general nature of viral en- 
cephalitis in the United Kingdom. In: lUis LS, ed. 
Viral Diseases of the Central Nervous System. 
Baltimore. WiUiams & Wilkins. pl9, 1975. 



Received January 14. 1982. P.S.E.B.M. 1982, Vol. 

171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 79-82 (1982) 



The Yucatan Miniature Swine: An Improved Pig Model for the Study of 
Desoxycorticosterone-Acetate (DOCA) and Aldosterone Hypertension (41480) 

JAMES M. TERRIS^ and RICHARD C. SIMMONDS 

Department of Physiology, Uniformed Services University of the Health Sciences, 
Bethesda, Maryland 20814 



Abstract. Blood pressure responses to chronic desoxycorticosterone-acetate (DOCA) or 
{/-aldosterone (Aldo) administration in the intact adult Yucatan miniature boar were investi- 
gated. Daily pressure measurements were made between 9:00 am and noon from Tygon 
carotid artery catheters. Following several days of baseline observations, DOCA-impregnated 
silastic strips, Aldo-impregnated silastic strips, or silastic alone (control) were implanted 
subcutaneously in the right or left flank under light thiamylal (Surital) anesthesia. Observa- 
tions were continued for 3 weeks. Blood pressures with either steroid were significantly 
different from control within one week postimplantation. They continued to rise throughout 
the study period from a preimplant level of 100- 1 10 mg Hg to a mean of 140 mm Hg. These 
studies demonstrate that the Yucatan miniature boar, in contrast to other animal models 
previously described, will readily develop hypertension with either DOCA or Aldo and that 
such studies can be conducted in the intact adult animal. 



It has previously been reported that 
young, uninephrectomized farm pigs re- 
spond to desoxycorticosterone-acetate 
(DOCA) implantation with a rapid and con- 
sistent rise in blood pressure (1, 2). While 
these animals responded more reliably than 
dogs and rats, the use of a rapidly growing, 
pediatric animal presented difficulty in the 
interpretation of data (1, 3). Since this 
model was really a characterization of 
juvenile hypertension, it was questioned 
whether the responses observed could be 
directly extrapolated to the adult hyperten- 
sive process. Adult domestic farm pigs can 
weigh 300 kg or more. Studies utilizing 
these animals in their adult stage were 
therefore not feasible. The present study 
was undertaken to determine whether the 
adult Yucatan miniature swine would be 
suitable for studies in this area. Mature 
Yucatan miniature boars have a mean 
weight of 83 ± 12 kg (range 59- 105 kg) (4) 
and, when treated correctly, are extremely 
docile. These features of the breed permit 



* To whom all correspondence should be addressed: 
Department of Physiology. Uniformed Services Uni- 
versity of the Health Sciences, 4301 Jones Bridge 
Road, Bethesda, Md. 20814. 



measurements to be made in the unre- 
strained adult animal, and, since it has been 
shown that uninephrectomy is not required 
(5, 6), hypertension studies can be con- 
ducted in the intact animal. Additionally, it 
was of interest to determine whether the 
intact adult animal would respond to the 
administration of J- aldosterone (Aldo) as 
well as DOCA with an increase in blood 
pressure and whether the responses would 
be similar. Significant aldosterone hyper- 
tension in an intact adult animal model has 
not been previously reported. 

Materials and Methods. Experimental 
animals (18-28 months old, 70-100 kg), 
were housed in 4 x 6-ft metabolic cages for 
observation periods of 2-4 weeks before 
undergoing surgical procedures. Through- 
out the study all animals received a premea- 
sured quantity of pig chow meal (Zeigler 
Brothers, Gardners, Pa.) supplemented 
with sodium chloride to allow a controlled 
sodium intake of 4.5 meq/kg/day. Water 
was provided ad libitum. All metabolic 
measurements had stabilized by the end of 
the observation period. 

An indwelling carotid artery catheter 
(Tygon 0.040-in. i.d., 0.070-in. o.d.) was 
then placed in each experimental animal. 
Following preanesthesia with ketamine (20 



79 
0037-9727/82/090079-04SQ V .^^ 

All hghl% T«»erv«d. 



> ! .li. f-rr I . I j:'.»'\.z"^'s-*.' — * "irr 
» -■;:■: I. -in ' •.•.." 7Pr:<:irc i7TT-r 
r. •'. "^rr- 'iT.-r"-^ •^'*r-ii: "!£■ Ji?lX 'I 

ni ;*-i !i:". T '.ir .. rfTlV I ..IT J n ■ ?ni^XISii 

ii: -r :'• .i:uti! ■.ur^'«ic. -.-.jd tt- 
i'.nT-..i:.i;i - !. 1 .•.nmt: 1 01 :r:tr 

.r f -».t:.*'. r"-*. ' r.i: " "COmii. 1* 7.i: 
-. .n ..'«:•■ . " t?*- TLrTHi-f;!:. : divi^ : : 

. .-. . . ...^ • . I* - -'.-.r.' .: . ."• ■.T' .'.. 

, . I .- ' .\ ■ .J." . .*- * .-T" •.: ■" "•T'.jrri!' T rr . l i '" i 

/ / ,. / .1. -.• . . .-- r. -- .; .*~ : '. " rrr* "•^fr■c: • 



- 


t » 


. \rfi 1 


.1 


.•t 




' 


. ■- 


i'.fl 


• • .» 




> ■ « 




-,.,'. 




.1 


I-'-. 


.* ,• 1 




»* - 


r r 




.■.*..' 




n.i 



/ / 



f ^•.- . #' ^ .. • . •- -r ^ -r - — :": — -j^: i 11.-= . ITc TlCii 

. '.,..• ••■•:;* 'i.TM: JTvi -. n^i.^'^iim:' 

/ . .••'.- ^'.»' - t- . .:. .I'T.^i ::e'%t aumui^ 

... // . / .■• ' '.: '. v-- * "4 L:^.:'i.':i±- ri- 

■'.••.: • : '-. . > : i r :i-T r»i 

, ^ •-. 1 * 111 .' '".i-. . "T . rT«iT'Tl''ivl - 

/ ... - / .•-•..'. -.-• " • i.-.r^ _-: :riiir 

/ . . .• • • \'\ ' ■' • ' : • Z'-t'.r :z :: r*t 

. • / • ■ . . "... '. t- : : .'i'ii «-.r 

^ •■ .-■/' • •'■,.'. : • ,^. : .'t .1' ■'t't'':'t r^ 

I* . .' • • ■ /' ;■■'■."-:' 1 '■ jt.' ^ r ..i^-ng ir.e 

.-•■;.■• - -. •.• .:•-.'.: "■:;. >' : :hi: :he re- 

, . , : : ,. ' • :• ' ;■•'■ ' '/••'■•.;.-•*: r' > i'^t k:dne> lo 

'/, '/', • '.■.'.''. t.f .f f/'/' /• .. '3', •■ r*. ■;!?•;•:■ > !!• 

.. , . ; ..•.,••«. / i.j'. /■.■!.•,■. i'K Ai'i'i r. .J . r;';.:ri :jsed in a number 

,11,*,* .. , J . . 1 ...... I tt,t '.ui'd'.ui <.l Mj.iji . /,iif, ro?'. The ;ihilit\ of this 

ii . .1 , .. I .. I,., ii..' ..I • . Ml i'l< h m.mI h. pMMlij'..*- [jypcrtcnsion has been 

I,. It I, I I I r iii|i' i' t ifi'i till 'li ■I'liiiintiri}' ;iii(l ihc results have been 



HYPERTENSIVE YUCATAN MINIATURE PIG 



81 




IMPLANT OUT 



I I 10 12 14 10 II 20 

DAYS 

Fig. 1. Average mean arterial pressure of control, DOCA-hypertensive, and J-aldosterone- 
hypertensive pigs from 5 days before until 20 days after implantation of silastic (control), DOC A 
silastic, or J- aldosterone silastic. Pressures were measured using indwelling catheters advanced into 
the thoracic aorta via the left or right carotid artery. Data are presented as mean ± SEM. Mean 
pressures increased regularly in the DOCA- and J-aldosterone-implanted animals, and were signifi- 
cantly different from control by 4 days postimplantation with </- aldosterone (*) and 7 days postimplan- 
tation with DOCA (*). 



conflicting. It is possible that the use of the 
racemic mixture, </,/- aldosterone, rather 
than the {/-isomer alone may have been the 
source of some of the problems encoun- 
tered in attempting to produce Aldo hyper- 
tension in an animal model (12). In several 
cases the isomer used is not stated. In addi- 
tion, the number of daily injections, vehicle 
used, and doses administered diflfer consid- 
erably from one study to another. The pres- 
ent studies demonstrate than an intact adult 
animal will respond to the administration of 
Aldo as well as DOCA with a rapid and 
consistent increase in blood pressure. The 
Yucatan miniature boar therefore provides 
an excellent opportunity for comparative 
studies of the actions of DOCA and Aldo in 
an intact animal in its adult stage. 



It has been reported that captopril, an in- 
hibitor of kininase II (angiotensin-I-con- 
verting enzyme) had no effect on the de- 
velopment of DOCA -salt hypertension 
or on established DOCA -salt hypertension 
when administered in the drinking water of 
rats (13, 14). However, the same inves- 
tigators subsequently reported a decrease 
in blood pressure in established aldoste- 
rone- salt hypertension and an attenuation 
of the hypertension when captopril was 
given during the developmental phase (IS). 
These findings imply mechanistic differ- 
ences in the hypertensive process resulting 
from the administration of these two 
steroids. The authors speculated a differ- 
ence in the importance of the vasodilator 
kinins. Vascular reactivity studies in the rat 









• ■"*— ~"^. -M*' 

~ r 1 "_ ■ r«r" -' t ; r kUL . in**" »■: 



:d .Tnrrnij 



. • r >- 



^vT 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 83-87 (1982) 



Concomitant Enhancement of B-Cell Mitogenesis and Inhibition of Antibody 
Synthesis by a Phorbol Ester (41481) 

THOMAS A. FERGUSON, LARRY A. FISH, C. STUART BAXTER, and 

J. GABRIEL MICHAEL' 

Departments of Microbiology and Environmental Health, University of Cincinnati College of Medicine, 

Cincinnati, Ohio 45267 



Abstract. 12-0-Tetradecaroyl-phorbol-13-acetate (TPA), a potent tumor-promoting 
agent, was found to enhance proliferation of murine splenic lymphocytes in response to 
B-cell mitogens, while possessing no mitogenic activity by itself. At the same time, TPA 
inhibited the B mitogen-induced polyclonal responses as well as antigen-specific antibody 
responses of cultured murine spleen cells. Our results support the hypothesis that tumor 
promoters inhibit lymphocyte differentiation. 



12-O-tetradecanoyl - phorbol - 13 - acetate 
(TPA) is a potent tumor-promoting agent 
which enhances the oncogenic effect of 
carcinogens on mouse skin. To learn the 
basis of its tumor-promoting activity, the 
morphological and biochemical responses 
of cell cultures treated with TPA have been 
the subject of intense investigations. Early 
reports by Sivak and Van Duuren (1,2) re- 
ported that the initial binding site of TPA is 
the cell membrane. Driedger and Blumberg 
(3) have shown that specific membrane 
binding activity for a series of phorbol es- 
ters correlates with their in vivo promoting 
activity. Subsequent to interaction with a 
membrane receptor TPA induces a series of 
biochemical events indicative of rapid cell 
proliferation that include increases in phos- 
pholipid synthesis and metabolism (4, S), 
RNA and DNA synthesis (6, 7), polyamine 
synthesis through stimulation of ornithine 
decarboxylase activity (8), and uptake and 
transport of amino acids (9). TPA is 
mitogenic for a variety of cell types, in- 
cluding 3T3 fibroblasts (10), chick embryo 
myoblasts (11), and chondroblasts (12), 
hamster embryo fibroblasts (13), and 
peripheral blood lymphocytes obtained 
from primates (14). Proliferative effects of 
the T-lymphocyte mitogens concanavalin A 
and phytohemagglutinin on lymphocytes 



' To whom reprint requests should be addressed. 



from bovine (15) and guinea pig (16) lymph 
nodes, and mouse spleens (17) were shown 
to be enhanced by TPA. 

TPAs mitogenic properties have been 
linked to its ability to inhibit cell differ- 
entiation by possibly committing cells to 
continuous replication (18). Cohen et al. 
(19) showed that TPA prevents terminal 
differentiation of chick embryo myoblasts 
in vitro, and Yamasaki et al. (20) and Ro- 
vera et al, (21) showed similar effects on 
Friend erythroleukemia cells. It has also 
been shown that promoters inhibit mor- 
phological differentiation of cultured neu- 
roblastoma cells (22). Contrary to the above 
reports several workers found that TPA 
may induce cell differentiation in erythroid 
cells of human promyelocytic leukemia 
cells (23-25). 

In the present study we investigated the 
effects of TPA on in vitro lymphocyte re- 
sponses utilizing assays which measure 
proliferation (mitogenesis) and terminal 
differentiation (antibody formation). We 
report that TPA is comitogenic when com- 
bined with the B-cell mitogens bacterial 
lipopolysaccharide (LPS) and muramyl di- 
peptide (MDP), while it inhibits the poly- 
clonal antibody response induced by the 
same mitogens, as well as the antigen- 
specific response to sheep red blood cells 
(SRBC). These data support the hypothesis 
that TPA prevents terminal differentiation in 
uncommitted lymphocytes. 



83 

0037-9727/82/09Q«^^-^S%^\ .^5W^ ^^_^ 

Copyright (g) \W1 b^ tbe Soc\«l^ \o« ^xvwvs«m&A^>s3«j«i %aA\*««KM» 
All rights reserved. 



84 



IMMUNOLOGICAL EFFECTS OF PHORBOL ESTERS 



Materials and Methods. Animals. BDFi 
(C57 Bl/6 X DBA/2) female mice 6-9 
weeks of age were purchased from Jackson 
Laboratories, Bar Harbor, Maine. 

Mitogens, antigens, and reagents. The 
phorbol ester, 12-O-tetradecanoyl-phorbol- 
13-acetate (TPA), was obtained from D and 
L Biochemicals, Milwaukee, Wisconsin, 
and dissolved in acetone at a concentration 
of 100 /xg/ml. Before use, TPA was diluted 
in culture media to the appropriate concen- 
tration. B-CeU mitogens used were bacterial 
lipopolysaccharide (£. coli 0127:08, Difco 
Labs., Detroit, Mich.) and muramyl di- 
peptide (N-acetylmuramyl-L-alanyl-D-iso- 
glutamine, MDP, a gift of Dr. Gordon 
Jones, Syntex Corp., Palo Alto, Calif.) 
Sheep red blood cells (Gibco, Grand Island, 
N.Y.) were washed three times with media 
and suspended to a concentration of 1% for 
culture immunization and 20% for plaque- 
forming cell assays. Results for mitogenic 
assays are expressed as counts per minute 
pH]thymidine incorporation. 

Culture media. Cell cultures were per- 
formed in RPMI 1640 containing 50 /xg/ml 
gentamycin, 2 x 10~* M 2-mercapto- 
ethanol, bicarbonate buffer, L-glutamine, 
and 10% fetal calf serum. 

Mitogenic assays. In vitro mitogenic as- 
says were performed on splenic lympho- 
cytes from BDFi mice. Cells (5 x 10^) were 
incubated in 0.2 ml culture media in 96-well 
microculture plates (Costar, Cambridge, 
Mass.) for 3 days with mitogen and/or TPA 
at 37° in a humidified atmosphere contain- 
ing 5% CO2. For the final 18 hr of culture 1 
fiCi pH]thymidine was added per well. 
Cells were harvested on glass fiber filters 
using a Skatron Cell Harvester (Flow 
Labs., Rockville, Md.) and pHJthymidine 
uptake determined by liquid scintillation 
counting. 

Plaque-forming cell cultures. In vitro 
cultures for antibody-forming cells were 
performed by the procedure of Mischell and 
Dutton (30). Splenic lymphocytes (5 x 
10«/0.5 ml) were incubated with SRBC (0.05 
ml of a 1% solution) or polyclonal activator 
(LPS or MDP) for 4 days in 24-well culture 
dishes (Costar) on a rocking platform under 
a gas mixture of 83% N2, 10% CO2, and 7% 



r* <N ^ r>i 

>C rsi ^ \p 

"O r* m -^ 

+1 +1 +1 +1 

5 o c* 8 

"tif ^ O <N 



Q 

Z 
< 

Q 

(/) 

z 
ttJ 
o 

g 



^ S r; J 



+1 +1 +1 +1 



lib 



U 
OQ 

Q 

Z 
< 

< 



Z 
u 
a 
o 

H 

i 
o 
u 



OQ 
< 



-a 

< 

Cl. 



^ s ^ 
<^ ^ — ;::; 

00 <N fO CO 
+1 +1 +1 +1 
O O >0 Q 
<^ ^ '- P 

>o o CO r* 



h 



r^ n o — 

vO ro — CO 

+1 +1 +1 +1 

vO >rj CO fS 

^ r- o ^ 

OS «o c^ «/^ 



'^ ^ — CO 
<N O <N O 

CM r- '^ in 

+1 +1 +1 +1 

— O — v-» 

o\ 10 vo r^ 
r>.^ "^^ *^ f^^ 

oC « '^' — 



E 



82 



Qu 

a 



CO ^ >0 

m ..i* t«« ^ 
— •^ Ov o 
\o r* CO <N 

+1 +1 +1 +1 

r>i w-i o r* 
r< OP iTi ^ 
r>i <N ^ r>i 



CO r>i 

r>i vo CO 

— fv| CO ^ 

vm 9m » CO 

•^ ^ >0 CO 

+1 +1 +1 +1 



•Ti CO '^ < 

00 00 v6 



•^ CO CO 



+1 +1 +1 +1 

>0 ^ OS <N 

r- -^ r^ ri 

CO vo iTi On 



§«.R- 



+1 +1 +1 +1 

<N r^ 00 CO 

r^ <N o CO 

«o ov r- '^ 



+1 +1 +1 +1 

<N — \0 CO 

<N r- o 1*^ 

CO «^ 00 00 

r^' co*^ trT <^' 
'^ ro <N 



82' 



C/5 

a. 



2 S 

8 E 

I 2 

1 ^ 

li S 

q3 <^ 

c3 +1 

2 c 

c .9 






o « J 



(5 E 



2> 



' o 



IMMUNOLOGICAL EFFECTS OF PHORBOL ESTERS 



85 



O2. Cells were fed daily with a nutritional 
cocktail (30). 

Direct plaque-forming cells. Direct PFCs 
were determined by the slide modification 
of the Jerne plaque assay (30). Cultured 
cells (0.05 ml) were mixed with 0.5 ml of 
0.6% agarose and 0.02 ml 20% SRBC and 
poured onto agarose-coated slides. Slides 
were incubated for 3 hr with 1/30 diluted 
guinea pig complement and the PFCs 
enumerated. 

Results. Table I illustrates that TPA en- 
hances cell division (as determined by up- 
take and incorporation of [^Hjthymidine) 
initiated by LPS or MDP at TPA doses of 
10"' to 1 /xg/ml. The synergistic effect be- 
tween LPS or MDP and TPA was evident if 
the reagents were added to the spleen cell 
cultures no longer than 6 hr apart, since we 
have found that if the interval was extended 
to 12 hr, the synergistic effects were not 
seen (data not shown). TPA was not 
mitogenic by itself at any dose tested, nor 
was any toxicity noted at the highest level 
used (1 /xg/ml). MDP alone as mitogen rou- 
tinely produced a stimulation of four to five 
times background while LPS produced a 
stimulation of 15-20 times. Because of 
TPAs marked effect on B-Cell proliferation, 
we wondered if the enhanced mitogenesis 
was reflected in an increased level of 
antibody synthesis. To measure TPAs ef- 
fect on the antigen-specific and polyclonal 



antibody induction, it was added to in vitro 
cultures of splenic lymphocytes and the 
polyclonal (LPS or MDP induced) or 
antigen-specific (SRBC immunized) anti- 
body synthesis was determined 4 days 
later using SRBCs as indicator cells. Re- 
sults in Table II demonstrate that TPA is a 
potent inhibitor of both polyclonal and 
antigen-specific antibody responses. To 
determine whether early or late events in 
lymphocyte response to antigen were af- 
fected, TPA was added to cultures at 24-hr 
intervals, each culture receiving addition of 
TPA on either Day 0, 1 , 2, 3, or 4 of culture. 
Results (Table III) indicate that addition of 
TPA to SRBC-immunized cultures from 
Days 0-3 significantly reduced the plaque- 
forming cell response to SRBC. TPA had 
no effect when added on Day 4, indicating 
that antibody secretion was not affected. 
Similar results were seen for the polyclonal- 
ly induced plaque-forming cell response 
with the suppression significant on Days 
Oand 1. 

Discussion. The dual effects of B lym- 
phocytes mitogens as activators of DNA 
synthesis and polyclonal antibody synthesis 
provided us with an opportunity to investi- 
gate the relationship between the require- 
ment for DNA synthesis and subsequent 
induction of polyclonal immunoglobulin 
production. In an earlier study we demon- 
strated that LPS is capable of activating re- 



TABLE IL Suppression of Antigen-Specific and Polyclonal Antibody Synthesis by TPA 



PFC/10« viable ccUs* 



Percentage inhibition 



Antigen specific 
SRBC" 
SRBC -h TPA' 

Polyclonal activation 
LPS" 

LPS + TPA 
MDP" 
MDP + TPA 

Controls 
TPA alone 
Media control 



406 ±26 

15 ±3 

191 ± 41 
110 ±31 
116 ±37 

10 ±3 

11 ±2 

16 ±6 



%.3'^ 

42.2'^ 
91.4^ 



" Cultures were inmiunized with 0.05 ml of a 1% solution of SRBCs. 

^ Direct PFCs ± SEM were determined using SRBCs as indicator cells. 

^ Significantly different from control as determined by Student's / test, P < 0.01. 

** LPS (10 fig/m\) or MDP (100 fig/ml) were added to cultures at initiation. 

' TPA was added to a final concentration of 1 fig/ml. 



86 



IMMUNOLOGICAL EFFECTS OF PHORBOL ESTERS 



TABLE III. Effect of Addttion of TPA 
AT Vauous Times during Culture 





PFC/10* viable ceDs* 


iMg/ml 

TPA added 

on Day:* 


SRBC 
specific'' 


polyclonal activator^ 
MDP LPS 




2 

3 

4 

None 


19 ±2 

28 ±1 

50± 10 

169 ± 10 

377 ±65 
341 ± 16 


11 ± 1 
21 ±5 
135 ±40 
153 ±28 
108 ±44 
105 ±44 


167 ± 19 
196 ±36 
265 ±25 
333 ± 10 
308 ±39 
465±20 



* Suppression significant when TPA added on Days 
0, I, 2. 3 to SRBC and Days 0, 1 in polyclonal activa- 
tion. /^ < 0.01. 

* Direct PFOIO^ viable ceDs ± SEM determined on 
Day 4 using SRBCs as indicator ceDs. 

*" Cultures were immunized with SRBCs on Day 0. 
^ MDP ( 100 Mg^ml) or LPS ( 10 /ig ml) was added on 
Day 0. 



sponsive B cells to polyclonal antibody 
production without significant cellular pro- 
liferation (26). Both hydroxyurea and 
cytosine aiabinoside, known inhibitors of 
DNA synthesis, failed to inhibit the LPS- 
induced polyclonal response, substantiating 
a dissociation between proliferative and 
polyclonal responses to LPS. In the present 
study addition of TPA to splenic cultures 
enhanced LPS and MDP mitogenicity but 
caused profound suppression of antibody 
synthesis. Our results support the hypothe- 
sis that TPA inhibits early stages of lym- 
phocyte differentiation into antibody- 
forming cells, as the most profound sup- 
pression of antibody synthesis was seen 
when TPA was added on Days and 1 of 
the culture. It appears unlikely that TPA is 
affecting the synthesis of products of the 
differentiated state (i.e., antibody), since 
TPA does not block antibody synthesis or 
secretion in murine myeloma cells which 
are already terminally differentiated ( 11 ) or 
affect antibody-forming cell cultures in 
these experiments when added on Day 4. 

It is conceivable that TPAs effects on 
lymphocyte differentiation contribute to its 
tumor-promoting activity. In support of this 
thesis. Keller (27) has shown that TPA in- 
hibits the killing of tumor cells by mac- 



rophages, and Mastro and Mueller (28) and 
Fish et aL (29) demonstiated inhibition of 
the mixed lymphocyte response by TPA. 

An alternative possibility would encom- 
pass the current knowledge of the effects 
of TPA on accessory cells (niacnH>hages). 
Mizel et al. (31) reported that TPA stimu- 
lates lymphocyte-activating £Eu:tor (LAF or 
Interieukin 1) production in the macrophage 
cell line P388Di, and Earrar et aL (32) 
showed enhanced production of Interieukin 
2 (1L2, formeriy T-cell growth fector) by T 
cells in the presence of phorbol ester. IL2 is 
known to result from ILl production from 
macrophages (33). It is conceivable that 
TPA stimulates these and other factors 
which enhance the mitogenesis of B lym- 
phocytes. The inhibition of antibody syn- 
thesis observed may be a secondary effect 
resulting from the activation suppressor 
cells by these macrophage and/or T-cell 
factors. Further work on the subcellular 
events involved will answer these ques- 
tions. 



1 . Sivak A, M ossman BT, Van Duuren BL. Activa- 
tion of ceD membrane enzymes in Uie stimulation 
of all division. Biochem Biophys Res Commun 
46:605, 1972. 

2. Sivak A. Raym F. Van Duuren BL. Phoiix>l ester 
tumor promoting agents and membrane stability. 
Cancer Res 29:624. 1969. 

3. Oriedger PE. Blumberg P. Specific binding of 
phorbol ester tumor promoters. Proc NaU Acad 
Sci USA 77:567, 1980. 

4. Levine L, Hassid A. Effects of phoibol-12.13 - 
diesters on prostaglandin activity in canine kidney 
(MDCK) cells. Biochem Biophys Res Commun 
79:477. 1977. 

5. Suss R. Kreibich G. Kinzel V. Phoibol esters as a 
tool in cell research. Eur J Cancer 8:299, 1972. 

6. Peterson .AR. Mondal S, Brankow DW, Thon W. 
Heidelberger C. Effects of promoters on DNA 
synthesis in C'H IDTl 2 mouse fibroblasts. Can- 
cer Res 37:3223. 19^:*. 

7. Suss R. Schuster .\. Tumor promoting Crotonoil 
factor tetradecano\l - phorbolacetate stimulates 
th>midine incorporation in normal and Con-A 
stimulated ih>mocyies. Experentia 30:81, 1974. 

8. OBrien TG. Simisman RC. and Boutuell RK. In- 
duction of the pol>amine-bios\nthetic enzymes in 
mouse epidermic b\ tumor promoting agents. 
Cancer Res 35:1662. 1975. 

9. Kensler TW . Weriz PW. and Mueller GC. Inhibi- 



IMMUNOLOGICAL EFFECTS OF PHORBOL ESTERS 



87 



tion of phorbol ester-accelerated amino acid 
transport in bovine lymphocytes. Biochim Phys 
Acta 585:43, 1979. 

10. Diamond L, O'Brien TG, Rovera G. Inhibition of 
adipose conversion of 3T3 fibroblasts by tumor 
promoters. Nature (London) 269:247, 1977. 

11. Diamond L, O'Brien TG, Rovera G. Tumor 
promoters: Effects on proliferation and differ- 
entiation of cells in culture. Life Sci 23, 1978. 

12. Pacifici M, Holtzer H. Effects of a tumor promot- 
ing agent on chondrogenesis. Amer J Anat 
150:207, 1977. 

13. Lankas GR Jr, Baxter OS, Christian R. Effect of 
tumor promoting agents on mutation frequencies 
in cultural V79 Chinese hamster cells. Mutat Res 
45:153, 1977. 

14. Abb J, Bayliss G, Deinhardt F. Lymphocyte acti- 
vation by the tumor promoting agent 12-O-tetra- 
decanoyl-pherbol- 13-astate (TPA). J Immunol 
122:1639, 1979. 

15. Mastro A, Mueller GC. Synergistic action of 
phorbol esters in mitogen-activated lymphocytes. 
Exp CeU Res 88:40, 1974. 

16. Rosenstreich DL, Mizel S. Signal requirements 
for T-lymphocyte activation. 1. Replacement of 
macrophage function with phorbol myristate ace- 
tate. J Immunol 123:1749, 1979. 

17. Fish LA, Baxter CS, Bash J A. Comitogenicity of 
tumor promoting. J Toxicol AppI Pharmacol 
48:A151, 1979. 

18. Boutwell RK. The function and metabolism of 
promoters of carcinogenesis. CRC Crit Rev Tox- 
icol 2:419, 1974. 

19. Cohen R, Pacifici M, Rubinstein N, Biehl J, Holt- 
zer H. Effect of tumor promoters on myogenesis. 
Nature (London) 266:538, 1977. 

20. Yamasaki H, Fibach E, Nudel U, Weinstein IB, 
Rifkind RA, Marks PA. Tumor promoters inhibit 
spontaneous and induced differentiation of murine 
erythroleukemia ceOs in cukure. Proc Natl Acad 
Sci USA 74:3451, 1977. 

21. Rovera G, O'Brien TG, Diamond L. Tumor 
promoters inhibit spontaneous differentiation of 
Friend erythroleukemia cells in culture. Proc Natl 
Acad Sci USA 74:2894, 1977. 

22. Ishii D, Fibach E, Yamasaki H, Weinstein IB. 
Tumor promoters inhibit morphological differ- 



entiation in cultured mouse neuroblastoma cells. 
Science 200:556, 1978. 

23. Rovera G, O'Brien TG, Diamond L. Induction of 
differentiation in human promyelocytic leukemic 
cells by tumor promoters. Science 204:868, 1979. 

24. Miao R, Fieldsteel A, Fodge D. Opposing effects 
of tumor promoters on erythroid differentiation. 
Nature (London) 274:271, 1978. 

25. Huberman E, Callaham M. Induction of terminal 
differentiation in human promyelocytic leukemia 
cells by tumor promoting agents. Proc Natl Acad 
Sci USA 76:1293, 1979. 

26. Poe W, Michael J. Separation of the mitogenic and 
antigenic responses to bacterial lipopolysac- 
charide. Immunology 30:241, 1976. 

27. Keller R. Suppression of anti-tumor defense 
mechanisms by phorbol esters. Nature (London) 
282:729, 1979. 

28. Mastro A, Mueller GC. Inhibition of the mixed 
lymphocyte proliferative response by phorbol es- 
ters. Biochim Biophys Acta 517:246, 1978. 

29. Fish LA, Baxter CS, Bash JA. Parallel order of 
reactivity on murine cells mezereine and phor- 
bol esters toward the mixed lymphocyte response 
and promotion of tumor regenesis. Toxicol Appl 
Pharmacol 59:173-176, 1981. 

30. Mischell R, Dutton R. Immunization of dis- 
sociated spleen cell cultures from normal mice. J 
Exp Med 126:423, 1%7. 

31. Mizel SB, Rosenstreich DL, Oppenheim JJ. Pher- 
bol myristate acetate stimulates LAF production 
by the macrophage cell line P388D. Cell Immunol 
40:320, 1978. 

32. Farrar JJ, Mizel SB, Fuller-Farrar J, Farrar WL, 
Hilfilter MI. Macrophage-independent activation 
of helper T-cells. I. Production of Interieukin 2. J 
Immunol 125:793, 1980. 

33. Oppenheim JJ, Mizel SB. Melter MS. Biological 
effects of lymphocyte and macrophage-derived 
mitogenic amplification factors. In Cohen S, Pick 
E, Oppenheim JJ. eds. Biology of the Lympho- 
kines. New York, Academic Press, p291, 1979. 



Received December I ] 
171. 



1981. P.S.E.B.M. 1982, Vol. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 88-91 (1982) 

Cobalamin (Vitamin B^z) Analogs Are Absent in Plasma of Fruit Bats Exposed to 

Nitrous Oxide' (41482) 

J. VAN DER WESTHUYZEN,* F. FERNANDES-COSTA,* J. METZ,* 
S. KANAZAWA,t G. DRIVAS,t and V. HERBERTt* 

^Department of Hematology, School of Pathology, South African Institute for Medical Research, and the 

University of the Witwatersrand, Johannesburg, 2000 South Africa, ^Hematology and Nutrition Laboratory, 

Veterans Administration Medical Center, Bronx, New York 10468, and the State University of 

New York Downstate Medical Center, Brooklyn. New York 11209 

Abstract. Fruit bats are an animal model for the neurologic damage which occurs in 
vitamin B,z-deficient humans. Cobalamin (vitamin B,}) analogs were not detected in the 
plasma of fruit bats treated with nitrous oxide (N2O), which inactivates cobalamin. This 
observation does not lend support to the suggestion that the neurological changes associated 
with cobalamin inactivation by NxO and/or cobalamin deficiency per se may be related to the 
accumulation of cobalamin analogs. However, although the plasma of control fruit bats 
lacked analogs, we did fmd analogs in their livers, at levels about 10% of total liver 
corrinoids, similar to human liver analog levels. 



It was recently suggested that physio- 
logically inactive cobalamin (vitamin B12) 
analogs are present in mammalian plasma 
(1) and tissues (2). If present, these analogs 
may have considerable importance, for 
they could exacerbate the effects of true 
cobalamin deficiency (3), such as the 
neurological changes. Kondo et al. (4) re- 
ported that exposure of rats to nitrous oxide 
(N2O), which inactivates cobalamin (5), 
also results in the conversion of cobalamin 
to analogs in the liver, but that the animals 
do not develop neurological changes. When 
fruit bats are fed a cobalamin-free diet, they 
develop neurological changes similar to 
those seen in human cobalamin deficiency, 
within 9 to 12 months (6), but this is not 
associated with the presence of cobalamin 
analogs (7). As gross neurological changes 
occur rapidly in fruit bats exposed to N2O 
(our unpublished data), the purpose of the 
present study was to determine whether 
cobalamin analogs played a role in the de- 
velopment of the N20-induced neurological 
changes in these animals. 

Materials and Methods. Experimental 



' Supported by grants from the South African 
Medical Research Council, the Research Service of 
the Veterans Administration, and USPHS Grant 
AM20526. 

'^ Vo whom correspondence should be addressed. 



animals. Fruit bats (Rousettus aegyptiacus) 
were captured in the wild and rendered 
cobalamin-deficient on an all-fruit diet (6). 
To prevent other vitamin deficiencies from 
developing, 0.2 ml of a cobalamin-free oral 
vitamin preparation (Abidec, Parke-Davis) 
was administered every 2 weeks in a dose 
containing 100 lU vitamin D, 0.2S mg 
thiamine, 0.01 mg riboflavin, 0.12 mg 
niacin, and 12.5 mg ascorbic acid. Bats 
maintained on this diet became cobalamin- 
deficient after 9 to 12 months. Control bats 
received intramuscular injection of O.S /xg 
cyanocobalamin per 100 g body weight 
every 2 weeks. 

Exposure to nitrous oxide. Bats were ex- 
posed to an atmosphere of 50% 02^50% 
N2O for 90 min every day for 3 weeks in a 
specially constructed cabinet in which CO2 
and water vapor were controlled. For 22.5 
hr a day, the bats breathed room air in their 
usual aviary, in which they had ample room 
to fly. 

Measurement of cobalamin analogs. 
Blood was drawn by cardiac puncture into 
heparinized tubes. The plasma was sepa- 
rated by centrifugation and stored at -20**. 
Plasma cobalamin analogs were determined 
by the Lau et al. coated charcoal radio- 
isotope dilution technique (8) as mod- 
ified by Kolhouse et al. (1), using salivary 
R-binder as ligand to measure total cor- 
rinoids, and pure intrinsic factor (IF) as 



88 
0037-9727 82 (y90088-04$01 .00/0 

Copynicht <; 1982 by the Society for ExperimentaJ Biology and Medicine. 
All nghts reserved 



B„ STATE OF FRUIT BATS ON NITROUS OXIDE 



89 



ligand to measure intact cobalamins, based 
on the facts that IF has a very low affinity 
for cobalamin analogs other than intact 
cobalamins, whereas the affinity of R- 
binders for these analogs is so much greater 
that it binds the totality of intact cobalamins 
plus other coninoids (1, 9). 

The R'binder used was human saliva, and 
the IF used was a gift from Becton- 
Dickinson Immunodiagnostics (Orange- 
burg, N.Y.) of pure hog IF prepared by 
affinity chromatography. The process for 
extraction from serum of cobalamin and 
analogs was that used by Kolhouse et al. 
(1), as was the amount of cyanide and 
BSA (bovine serum albumin) added. Sep- 
aration of bound from free cobalamin was 
with albumin-coated charcoal prepared as 
described by Lau et al, (8). The reference 
standard was USP (United States Phar- 
macopeia) Cyanocobalamin Standard, pur- 
chased from the USP (Washington, D.C.). 
We have used this identical methodology to 
study cobalamin and analog levels in human 
serum, red cells, brain, liver, and bile 
(10-12). The amount of analogs present is 
thus represented by subtracting from the 
total coninoids detected when R-binder is 
used as ligand at pH 9, the quantity of intact 
cobalamins measured at pH 9 with IF as 
ligand (1). 

Results. The results are shown in Table I. 



The group of four cobalamin-replete fruit 
bats received intramuscular injections of 
cyanocobalamin every 2 weeks during the 2 
months they were in captivity. Serum 
cobalamin levels were normal, varying 
from 1536 to 2781 pg/ml when assayed with 
IF as ligand. No significant amounts of 
cobalamin analog were present in the 
plasma of these bats, the analog ranging 
from to 126 pg/ml. 

The bats rendered cobalamin deficient 
had received the cobalamin-free diet for 12 
to 21 months and all were severely cobala- 
min deficient (plasma cobalamin = 1-91 
pg/ml). No significant amounts of cobala- 
min analog could be detected in deficient 
animals exposed to N2O (range = 0-18 
pg/ml), or those not so exposed (range = 
0-33 pg/ml). 

Discussion. In all fruit bats studied, 
cobalamin analogs were absent from the 
plasma or present in negligible amounts 
only. In the bats rendered cobalamin defi- 
cient by dietary means only, the results 
were similar to those reported by Green and 
Jacobsen (7). Cobalamin analogs were also 
not detected in the plasma of bats exposed 
to N2O, which suggests that analogs do not 
play a role in the development of the severe 
neurological changes that accompany N2O 
exposure in the bat. Furthermore, cobala- 
min analogs were detected in small amounts 



TABLE I. Plasma Levels of Cobalamins and Analogs (pg/ml) in Fruft Bats" 



Group. 


No. 


Cobalamins 


Total corrinoids 


Analogs 


Cobalainin-replete 


1 


2610 


2713 


103 




2 


2781 


2786 


5 




3 


1536 


1490 







4 


2747 


2873 


126 


Cobalamin-deficient 


5 


19 










6 


26 


59 


33 




7 


16 










8 


58 


77 


19 


Cobalamin-deficient 


9 


1 








exposed to NjO 


10 


23 


41 


18 




11 


91 


60 







12 


26 


1 






" Assayed by radioisotope dilution assays using as ligand pure intrinsic factor (IF) for cobalamins and salivary 
R-binder for total corrinoids (cobalamins + analogs). 



90 



B„ STATE OF FRUIT BATS ON NITROUS OXIDE 



in only two of four fruit bats cobalamin re- 
plete from intramuscular cobalamin injec- 
tions every 2 weeks; this slightly differs from 
the fmding by Green et al, (13) of quantity 
of analogs present in serum of fruit bats in- 
jected with 100 ng cyanocobalamin weekly 
(but not in other fruit bats). 

It is not clear why cobalamin analogs ap- 
pear in the tissues of NzO-treated rats (4) 
but not in the plasma of bats. This may rep- 
resent a species difference, for there is evi- 
dence that the bat and the rat respond dif- 
ferently to N2O exposure: N2O causes se- 
vere neurological changes in the bat but 
none in the rat, and the deoxyuridine (dU) 
suppression test (14) is abnormal in the 
N20-exposed rat (15) but not in the bat (our 
unpublished observations). Another possi- 
bility is that the length of exposure to N2O 
may be critical, for Kondo and co-workers 
(4) exposed rats continuously to N2O for 
periods ranging from 30 min to 38 hr, while 
the bats in the present study were exposed 
for 90 min daily for 3 weeks. It is also pos- 
sible that analogs present in tissues may be 
cleared more rapidly from the plasma of the 
bat than of the rat. 

The term '^analog" as used in the present 
study does not refer to any specifically iso- 
lated form or forms of corrinoids, but is 
used to describe molecules which have a 
corrin nucleus and therefore attach to R- 
binders as does vitamin B12, but in addition 
have a lesser affinity than cyanocobalamin 
for IF and/or a much greater affinity than 
cyanocobalamin for R-binder. 

When we studied the livers of 4 control 
and 6 nitrous-oxide treated bats, we found 
the control bats had a mean of 65.8 ± 16.1 
ng total corrinoid/g liver, a mean of 59.9 ± 
15.7 ng cobalamin/g liver, and a mean of 
5.9 ± 2 ng analog/g liver. Thus, although 
normal fruit bats lack cobalamin analogs in 
their serum, such analogs are present in 
their livers, suggesting more rapid clearance 
of analogs from serum into liver in bats than 
in humans (10- 12). Livers from the 6 NgO- 
treated bats had a mean of 1.3 ± 1.3 ng 
total corrinoid/g liver, a mean of 1.5 ± 0.5 
ng cobalamin/g liver, and no measurable 
analog in their livers. 



1. Kolhouse JF, Kondo H, Allen NC, P6deU ER, 
Allen RH. Cobalamin analogues are present in 
human plasma and can mask cobalamin defk:ieiicy 
because current radioisotopic dilution assays are 
not specific for true cobalamin. New Engl J Med 
299:785-792, 1978. 

2. Kondo H, Kolhouse JF, Allen RD. Presence of 
cobalamin analogues in animal tissues. Proc Nat 
Acad Sci USA 77:817-821, 1980. 

3. Kolhouse JF, Allen RH. Absorption, plasma 
transport, and cellular retention of cobalamin 
analogues in the rabbit. J Clin Invest 60:1381- 
1392, 1977. 

4. Kondo H, Osborne ML, Kolhouse JF, Binder MJ, 
Podell ER, Utley CS, Abrams RS, Allen RH. 
Nitrous oxide has multiple deleterious effects on 
cobalamin metabolism and causes decreases in 
activities of both mammalian cobalamin- 
dependent enzymes in rats. J Clin Invest 
67:1270-1283, 1981. 

5. Blackburn R, Kyaw M, Swallow AJ. Reaction of 
Cob(I)alamin with nitrous oxide and Cob(IIl)ala- 
min. J Chem Soc, Faraday Trans 73:250-255, 
1977. 

6. Green R, van Tonder SV, Oettle GJ, Cole G, Metz 
J. Neurological changes in fruit bats deficient in 
vitamin B,2. Nature (London) 254:148- 150, 1975. 

7. Green R, Jacobsen DW. No discrepancy between 
vitamin B12 radioassays using purified intrinsic 
factor or R binder in bats developing cobalamin 
deficient neuropathy. Clin Res 28:74A, 1980. 

8. Lau KS, Gottlieb C, Wasserman LR, Herbert V. 
Measurement of serum vitamin B12 level using 
radioisotope dilution and coated charcoal. Blood 
26:202-214, 1965. 

9. Gottlieb CW, Relief FP, Herbert V. Blockade of 
vitamin B,2-binding sites in gastric juice, serum 
and saliva by analogues and derivatives of vitamin 
B,2 and by antibody to intrinsic factor. Biochim 
Biophys Acta 141:560-572, 1967. 

10. Kanazawa S, Herbert V. Vitamin B,2 analog con- 
tent in human red cells. liver, and brain. Clin 
Res 30:504A, 1982. 

11. Herbert V, Kanazawa S, Drivas G, Foscaldi R, 
Manusselis C. Herzlich B, Colman N, van der 
Westhuyzen J, Fernandes-Costa F, Metz J. Evi- 
dence for presence of stable vitamin Bi2 analogs in 
serum of bushmen on "natural" diets, and in liver 
of Americans, and analog absence in red cells of 
Americans, and fruit bat serum (even after vita- 
min B,2 supplementation). In: Proceedings, 19th 
Congress of the International Society for 
Hematology, Budapest, August 1-7, p 236, 1982. 

12. Kanazawa S, Herzlich B, Drivas G, Manusselis C, 
Herbert V. Removal of cobalamin analog in bile 
may be the primary purpose of the enterohepatic 



B„ STATE OF FRUIT BATS ON NITROUS OXIDE 



91 



circulation of vitamin Bn. Clin Res 30:687A 
(1982). 

13. Green R, Jacobsen DW, Sommer C. Cobalamin 
depleted fruit bats show discrepancies between 
radioassays using purified intrinsic factor or R- 
binder after they are given a cyanocobalamin sup- 
plement. Blood 58 (Suppl. l):27a, 1981. 

14. Metz J, Kelly A, Swctt VC, Waxman S, Herbert 
V. Deranged DNA synthesis by bone marrow 



from vitamin Bi2-deficient humans. Brit J 
Haematol 14:575-592, 1968. 
15. Deacon R, Lumb M, Perry J, Chanarin I, Minty B, 
Halsey MJ, Nunn JF. Selective inactivation of 
vitamin B,2 in rats by nitrous oxide. Lancet 
2:1023-1024, 1978. 

Received January 26, 1982. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 92-97 (1982) 



Therapeutic Concentrations of Antineoplastic Agents Diminish Interferon 

Yields (41483) 

THOMAS C. CESARIO,* LEWIS M. SLATER, HAROLD S. KAPLAN, and 

JEREMIAH G. TILLES 

University of California, Irvine, California 92668 



Abstract. Interferon production was studied in the presence of therapeutic concentra- 
tions of five antineoplastic agents. None of the drugs studied grossly altered the production 
of interferon by fibroblasts. Virus-induced mononuclear cells, in contrast, produced signifi- 
cantly less interferon after prior exposure to vincristine and produced no detectable inter- 
feron after prior exposure to 6-MP. These studies establish that interferon yields m vitro may 
be diminished by therapeutic concentrations of antineoplastic agents, and necessitate in vivo 
studies to determine if this alteration in interferon production can help explain the increased 
severity of certain viral infections in patients with neoplastic diseases. 



Patients with certain forms of malignant 
disease are known to be at increased risk 
for serious viral infections (1). Because of 
this fact, we have questioned whether in- 
terferon production may be diminished in 
malignant disease particularly by the 
chemotherapeutic agents used to treat 
neoplastic processes. 

As a preliminary to investigations in vivo, 
we have studied the effects of various 
antineoplastic agents on the ability of 
human cells to produce interferon in vitro. 

This report reviews our experience with 
the production of interferon in the presence 
of five antineoplastic agents. We have 
found that therapeutic concentrations of 
these agents can influence interferon pro- 
duction in human cells, especially mononu- 
clear cells. We suggest in vivo studies be 
done to determine if this phenomenon oc- 
curs in cancer patients receiving chemo- 
therapy. 

Materials and Methods. Cells. Fibro- 
blasts. Human foreskin fibroblasts (FF 
cells) originally prepared in our own labo- 
ratory were used for these experiments. Fi- 
broblasts were grown to confluence in 25- 
cm^ plastic screw-cap flasks prior to inter- 
feron induction. 



' To whom correspondence should be addressed: 
Department of Medicine, University of California, 101 
City Drive South. Orange, Calif. 92668. 



Mononuclear leukocytes. Human leuko- 
cytes were obtained from the American 
Red Cross as the platelet-rich fraction of 
human blood. These preparations were 
tested within 24 hr of procurement. Upon 
receipt in the laboratory, the cell suspen- 
sion was subjected to centrifugation on 
Ficoll-Hypaque for 45 min at 1000 rpm. 
The mononuclear cell layer was then har- 
vested, washed three times with phos- 
phate-buflfered saline (PBS), and adjusted 
to a concentration of 1 x 10^ cells/ml 
in Newman Tytell medium containing 10% 
bovine fetal serum (BFS). 

Antineoplastic agents. Antineoplastic 
agents were obtained from the following 
sources: 5-fluorouracil from Roche Labo- 
ratories, Nutley, New Jersey; adriamycin 
from Farmitalia, SPS, Italy; vincristine 
from Eli Lilly, Indianapolis, Indiana; 6- 
mercaptopurine (6MP) from Burroughs 
Wellcome, Research Triangle Park, North 
Carolina; and methylprednisolone from 
Upjohn Company, Kalamazoo, Michigan. 

All solutions were freshly prepared im- 
mediately prior to use. The concentrations 
employed in these experiments were cho- 
sen to approximate those present in the 
serum immediately following administra- 
tion to humans (2-5). 

Inducing agents. Complexed polyinosin- 
ic-polycydidylic acids (I:C) were obtained 
either in liquid or powdered form and 
were diluted just prior to use in phosphate- 



92 
0037-9727/82/090092-06$0 1 .00/0 

Copyright ;,<:;^ 1982 by the Society for Experimental Biology and Medicine. 
' /ifh/s reserved. 



ANTINEOPLASTIC AGENTS AND INTERFERON 



93 



buffered saline (PBS) containing calcium 
and magnesium. 

Diethylaminoethyl dextran (DEAE-D) 
was dissolved in water, in a concentrated 
form, and frozen at -20**. Immediately 
prior to use, an aliquot of DEAE-D was 
added to the I:C solution so that the final 
concentration of the inducing solution was 
10 /Ltg/ml of I:C and 100 /xg/ml of DEAE-D. 

Newcastle disease virus (NDV) was 
grown in eggs using a laboratory strain that 
had been serially passed through many gen- 
erations. 

Induction of interferon. Preliminary ex- 
periments demonstrated that the various 
concentrations of drugs employed were not 
toxic to the fibroblast monolayers (as evi- 
denced by lack of morphologic changes) 
and would not induce interferon. 

FF cells were grown to confluence in 
screw-cap flasks. At the initiation of the 
experiment three flasks of fibroblasts were 
exposed to L-IS media containing 2% 
bovine fetal serum, 1 mg/ml dextrose, 30 
/Lig/ml glutamine, 90 fig/ml arginine, ISO 
u/ml penicillin, 250 /xg/ml of streptomycin, 
and an appropriate concentration of the 
antineoplastic agent to be tested. Control 
flasks and simultaneous exposure flasks 
were treated with this same media without 
the antineoplastic agent during this 24-hr 
period. After the preliminary exposure, I:C 
induction was accomplished by adding 2 ml 
of the inducing solution containing the indi- 
vidual antineoplastic agents in the concen- 
trations noted to both pretreatment and si- 
multaneous exposure flasks. Control flasks 
were provided with 2 ml of the inducing 
solution without the antineoplastic agents. 

All flasks were exposed to the inducing 
solutions for 3 hr and then washed with 
PBS. Subsequently, 2 ml of L- IS media 
supplemented as described above was 
added. Individual antineoplastic agents in 
the appropriate concentrations were added 
to the media of the test flasks (both pre- 
treated and simultaneously treated), but 
control flasks contained the same media 
without such agents. The flasks were incu- 
bated overnight and the media was then 
harvested. All solutions were dialyzed for 
24 hr against 100 vol of PBS to remove the 



antineoplastic agent, harvested, and frozen 
at -80** until assayed. Control and test in- 
ductions were performed simultaneously 
under identical circumstances save for the 
presence of the drug in the test solutions. 
All inductions were performed in triplicate. 

In the case of induction with virus, dupli- 
cate flasks were prior treated as described 
above. At the appropriate time, the prior- 
treated flasks and unexposed flasks were 
suctioned and exposed to NDV in a con- 
centration of 100 EID50 per cell for 1 hr. At 
the conclusion of this exposure, the two 
prior-treated flasks as well as the simulta- 
neous exposure flasks were covered with 2 
ml of the supplemented L-IS containing the 
antineoplastic agents. Duplicate control 
flasks, which had been exposed to virus but 
not to the test drugs, were covered with a 
similar quantity of supplemented L-IS 
without any antineoplastic drug. 

Incubation was then carried out for 24 hr 
after which the media was harvested and 
dialyzed against 100 vol of 0.1 A/ citric acid 
for 4 days. The dialysis bath was then 
changed to PBS for 24 hr. Subsequently, 
the solutions were harvested and ul- 
tracentrifuged at 25,000 rpm for 90 min. Su- 
pematants were harvested and stored at 
-80** until assayed. 

Induction of interferon in mononuclear 
leukocytes was accomplished as follows: 
cells were adjusted to a concentration of 10^ 
cells/ml in Newman Tytell media supple- 
mented with 10% BFS, 250 u/ml of peni- 
cillin, and 150 /xg/ml of streptomycin. 
Tubes used for the pretreatment experi- 
ments were exposed to media containing 
the appropriate concentration of antineo- 
plastic agent for 24 hr at 37** in a 5% 
CO2 incubator. Other tubes were simulta- 
neously incubated in identical media save 
for the absence of the drugs. 

Freshly prepared media was then added 
to all cells: the pretreatment and simuha- 
neous exposure tubes receiving media 
containing the appropriate drug and the 
control tubes receiving the media without 
the drugs. NDV in a concentration of 100 
EIDso/cell was added to each tube and in- 
cubation carried out for 24 hr in an in- 
cubator containing 5% CO2. At the conclu- 



¥4 



Al^flt^EOfLASTlC ACEXTS ASO MSTtMFEmom 



%km 44 the mctihsakm period the celb were 
cenuWutitd^ the wptrnsOMi huytutd^ 9od 
the rnedk^ dkAyzed stpanu ckric add tor 4 
dMy% Mid n% for I day. The media was 
then harveoed^ cemiifiiird for 90 mhi at 
25^M fpm, and frozen at -ttf for later 
a^ay. 

ViaMffty counts were performed by the 
trypan Mue exclusion method prior to dis- 
carding all celU. For all experiments at least 
tlit% of the celK were viable at the conclu- 
Mhn iff the experiment and the difference in 
viability between te^t and control tubes was 
always le^^ than 10%, 

Interferon uMMuyM. Interferon was as- 
sayed u%in§ a microtiter method which 
employed vehicular stomatitis virus as 
challenge and was performed on foreskin 
fibroblasts. This assay has been described 
prevhu%\y (6« 7) and includes a standard 



NIH fcf4 



assayed 
G4I23-901-527. 

RcMkiu Tables I and II describe oar le- 
softs. No differences were appreciated 
between the quantity ot interferao pro- 
duced by fibroblasts in the presence erf* the 
antineoplastic agents or in the absence of 
these same <inigs when I:C was used as the 
inducer. When Cbroblasts were induced 
using NDV, only pretreatment with adri- 
amycin was associated with any consis- 
tent reduction in interferon yield, but this 
reduction was of a minimal degree and its 
significance seemed slight. Methylpred- 
nisolone, SFU, 6MP, and vincristine had no 
adverse effect on virus-induced interferon 
in fibroblasts even with pretreatment. All 
resuhs were repeatedly reproduced in sepa- 
rate testing. 

In contrast to the results with fibroblasts. 



'I AMU', I, Tmi: HPHf t%o9' ANiiNhOFLAJnic: Agents ON Interferon Production in Fibroblasts 













Interferon titers 










Antineoplastic agent 




Antineoplastic agent 




Conccniralion 


Prw 

1 




present 








absent 






)r exposure 


Simultaneous exposure 


1 


2 




Agrnl 


2 


3 


1 


2 


3 


3 




l:C induction 












VimrlMirir 


((MW/ig/ml) 




















tril 1 




2(K) 


2(X) 


400 


400 


400 


400 


200 


200 


200 


Iril ; 




32(K) 


32(X) 


3200 


1600 


3200 


3200 


3200 


3200 


3200 


h Mrrviipldpiiiinc 


().2/ig/ml) 




















IcM 1 




2(X) 


2(X) 


200 


200 


400 


400 


400 


400 


400 


iVM : 




8(X) 


mo 


1600 


800 


800 


800 


800 


800 


1600 


AdiiuitiVi'in 


(0.40/ig/ml) 




















IVM 1 




8(K) 


1600 


1600 


400 


800 


800 


1600 


1600 


1600 


ivii ; 




2(X) 


2(X) 


200 


200 


200 


200 


200 


200 


200 


^ I'liioioiintv'il 


(40.0 /J It/ml) 




















lV*i( 1 




8(K) 


mo 


1600 


800 


800 


800 


800 


800 


800 


\V^H\ ) 




I MM) 


I6(M) 


1600 


400 


800 


1600 


400 


800 


800 


MrlhylpiiMlitlNolonc (KO.O /itt/ml) 




















lom 1 




2(K) 


4(X) 


4(X) 


2(H) 


400 


400 


400 


400 


400 


IrM .' 




4(H) 


8(H) 


8(H) 


4(H) 


400 


800 


400 


400 


800 








NDV inducti 


ion 












Vliuii^llnc 


lO(V8/iKml) 


MM) 


32(H) 


— 


32(H) 


32(X) 


— 


3200 


3200 


— 


^ Mouiiplopuimc 


M :/i)imh 


3:tH) 


3:(H) 


— 


16(H) 


32(H) 


— 


1600 


3200 


— 


Ailiituiis^ III 


l0 40^4Kml) 




















IrM 1 




INK) 


IWH) 


— 


8(H) 


32(X) 


— 


3200 


3200 


— 


loM : 




INK) 


IMH) 


— 


32(H) 


6400 


— 


6400 


6400 


— 


li'M \ 




OlH) 




— 


M(H) 


— 


— 


6400 


— 


— 


loM A 




IMH) 




— 


32(H) 


— 


— 


3200 


— 


— 


^ 1 tuoi\mi(K il 


l40 0M):ml> 


H1H> 


IMH) 


._ 


8(H) 


16(V 


— 


1600 


1600 


— 


NUMh\lpioJmsolv»no (SO ^«j; mh 


MIX) 


M(H) 


- 


M(H) 


64(H) 


— 


3200 


6400 


— 



ANTINEOPLASTIC AGENTS AND INTERFERON 



95 



TABLE II. The Effects of Antineoplastic Agents on Interferon Production 
IN Human Mononuclear Cells" 











Interferon titers 










Antineoplastic agent 




Antineoplastic agent 




Concentration 




present 




absent 




Prior 


exposure 


Simultaneous 


exposure 


1 




Agent 


1 


2 


I 


2 


2 


Vincristine 


(0.08 Mg/ml) 














Test 1 




100 


100 


400 


400 


800 


800 


Test 2 




400 


800 


800 


1600 


1600 


1600 


Tests 




200 


200 


— 


— 


800 


800 


6-Mercaptopurine 


(3.2 figfml) 














Test 1 




<100 


<100 


400 


800 


1600 


1600 


Test 2 




<100 


<100 


400 


800 


800 


800 


Tests 




<100 


<100 


— 


— 


1600 


1600 


Adriamycin 


(0.40 fig/m\) 














Test 1 




1600 


1600 


800 


1600 


1600 


1600 


Test 2 




800 


800 


400 


800 


800 


800 


5-Fluorouracil 


(40.0 Mg/ml) 














Test 1 




800 


1600 


1600 


1600 


1600 


1600 


Test 2 




800 


1600 


400 


800 


800 


800 


Methylprednisolone (80.0 fig/nH) 














Test 1 




400 


800 


400 


800 


800 


800 


Test 2 




800 


1600 


800 


1600 


1600 


1600 



' NDV induction. 



vincristine consistently lowered interferon 
yields when the leukocytes were pretreated 
with this drug, and 6MP, under the same 
conditions, virtually eliminated detectable 
interferon production. Both of the differ- 
ences were significant (P < 0.01 by Stu- 
dent's / test). Simultaneous exposure for 
both drugs resulted in a diminished yield in 
three of the four tests. Adriamycin, meth- 
ylprednisolone, and S-fluorouracil failed 
to diminish mononuclear cell interferon 
yields. Again the results were reproducible 
in separate testing. 

[Titration of NDV performed by hem- 
agglutination before and after induction 
in both fibroblasts and mononuclear cells 
demonstrated that no change in virus con- 
centration occurred during induction in 
either test or control flasks.] 

Discussion. This report demonstrated 
that antineoplastic agents can influence in- 
terferon production by human cells in vitro. 
The observed alterations in interferon pro- 
duction, however, related not only to the 
particular drug in question but also to the 
ceU type and the inducer. 



We found the production of fibroblast 
interferon after I:C induction was unaltered 
by any of the antineoplastic agents studied. 
Furthermore, while adriamycin appeared to 
diminish the yield of fibroblast interferon if 
the cells were pretreated with the drug for 
24 hr prior to induction with NDV, the dif- 
ference in interferon yield was minimal. 
None of the other drugs studied affected 
virus-induced interferon production by fi- 
broblasts. Thus, it seems fibroblasts were 
relatively resistant to the effects of the 
antineoplastic agents, at least as regards 
interferon production. 

In contrast to the observations made with 
fibroblasts, vincristine and 6-MP lowered 
the quantity of interferon produced by 
mononuclear cells after viral induction. The 
reduced yields, however, were consistently 
seen only when mononuclear cells were 
pretreated with the antineoplastic agent. 
Prior treatment with vincristine reduced the 
quantity of interferon produced by as much 
as 88% and pretreatment with 6-MP virtu- 
ally eliminated detectable interferon pro- 
duction. 



96 



ANTINEOPLASTIC AGENTS AND INTERFERON 



In the circumstances where diminished 
interferon yields were demonstrated the 
question arises as to whether these effects 
occurred because of alterations in the host 
cell or because of effects on the vims. Virus 
titrations were performed by hemaggluti- 
nation during induction and no viral prolif- 
eration could be documented under these 
circumstances. While infectivity measure- 
ments would have been a more reliable 
means of assessing viral proliferation, the 
fact that viral reproduction may not occur 
in this system suggests alterations in the 
cell are responsible for the observed dimi- 
nution in interferon yields. Furthermore, if 
the effects were on the virus it would have 
been expected that the drugs would have 
decreased interferon yields in both types of 
cells used rather than just one as we de- 
scribed. It therefore seems likely that the 
diminished yields were due to alterations in 
the cell induced. It is probable that cell 
macromolecular synthetic capacities were 
compromised in those circumstances where 
diminished yields were documented. This 
hypothesis will be verified in future studies. 
Furthermore, it is likely that pretreatment 
permits the compromise in synthetic func- 
tions to occur before interferon production 
begins while simultaneous treatment per- 
mits the initiation of interferon synthesis 
before the drugs can affect the cell. Previ- 
ous studies have provided a precedent for 
the assumption that protein-synthesizing 
capacities may be impaired at least by 6MP. 
Thus, Tidd and Paterson (8) have described 
alterations in RN A due to 6MP, and as both 
RNA (9, 10) and protein synthesis (11) are 
required for interferon production, chemi- 
cals which alter RNA could adversely in- 
fluence interferon yield. 

The fact that 6MP and vincristine altered 
production of interferon in mononuclear 
cells, but not fibroblasts, likely relates to 
the specific effects these drugs have on in- 
dividual cell types. Thus, 6MP affects 
primarily leukemic cells (12) and vincristine 
affects differing cell types including those 
derived from lymphoidal origin (12). 

The concentrations of chemotherapeutic 
agents used in these experiments were pur- 
posely selected to approximate the highest 
or peak therapeutic concentrations achieved 



in man. These high concentrations were 
used to maximize our opportunity to dem- 
onstrate any effect on interferon synthesis 
by these drugs. We have shown that at 
these peak therapeutic concentrations 6MP 
and especially vincristine can influence 
interferon production in mononuclear cells. 
Since, however, drug concentrations fall 
as the drug is eliminated by metabolism 
or excretion, it is now necessary to perform 
in vivo studies to find if the effect on inter- 
feron synthesis reported here can be dem- 
onstrated in the body. Further studies are 
currently in progress in our laboratory to 
find the minimum concentrations necessary 
to diminish interferon yields from mononu- 
clear cells and to determine the duration of 
drug exposure necessary to afifect this less- 
ened production of interferon. 

These studies offer some support to the 
hypothesis that the increased severity of 
viral infections occasionally observed dur- 
ing the course of neoplastic diseases could 
be at least in part due to diminished capac- 
ity of the patient to produce interferon in 
the presence of antineoplastic drugs. In 
vivo studies as mentioned above, however, 
will be necessary to confirm this hypoth- 
esis. 

While no other comprehensive study 
similar to this investigation has been carried 
out on human cells, other relevant investi- 
gations deserve comment. Kilboume et al. 
(13) using eggs, Postic et al, (14) using rab- 
bits, and Talas and Stoger (15) using mice, 
all demonstrated diminished interferon 
production in the presence of steroids. In 
contrast, I>eSomer et al. (16) working in 
rats found steroids failed to depress inter- 
feron levels and Mendelson et aL (17) found 
steroids actually increased interferon levels 
in mice. 

As regards studies in tissue culture mod- 
els, I>eMaeyer and DeMaeyer (18) working 
on rat tumor cells found steroids depressed 
interferon production, but Adolf and 
Swetly (19) have recently found that vari- 
ous steroids can augment interferon pro- 
duction of human lymphoblastoid cells 
when induced by viruses. 

Havell and Vilcek (20) have studied the 
effect of vinblastine, a drug related to vin- 
cristine, on interferon production and found 



ANTINEOPLASTIC AGENTS AND INTERFERON 



97 



little adverse influence; however, these 
authors noted this drug diminished inter- 
feron secretion. 

St. Geme et al, (21) investigated the ef- 
fects of 6MP on the ability of chick embryo 
fibroblasts to produce interferon, and while 
documenting resultant yields were slightly 
diminished, these authors felt the difTer- 
ences were not significant. 

The discrepancy between the results of 
some of these studies and our own are mul- 
tifaceted, but likely result from inherent 
dissimilarities between the animals used 
and between the basic cell types studied. 

1. Levine A, Schimpf S, Graw R, Young R. 
Hematologic malignancies and other marrow 
"Failure States: Progress in the management of 
complicating infections. Semin Hematol 11:141- 
201. 1974. 

2. Chabner B, Myers C, Olivera V. Clinical phar- 
macology of anticancer drugs. Semin Oncol 
4:165-191. 1977. 

3. Liddle G, Melmon K. The adrenals. In: Williams 
R, ed. Textbook of Endocrinology. Philadelphia, 
Saunders, pp233-322, 1974. 

4. Loo T, Luce J. Sullivan S, Frei E. Clinical phar- 
macological observations on 6-mercaptopurine 
and 6-methylthiopurine ribonucleoside. Clin 
Pharmacol Ther 9:180- 194, 1968. 

5. Owellen R, Root M, Mains F. Pharmacokinetics of 
vindisine and vincristine in humans. Cancer Res 
27:2603-2607, 1977. 

6. Cesario T, Schryer P. Tilles J. The relationship 
between the physicochemical nature of human 
interferon, the cell induced and the inducing 
agent. Antimicrob Agents Chemother 11:291- 
298, 1977. 

7. Tilles J, Finland M. Microassay for human and 
chick cell interferon. Appl Microbiol 16:1706- 
1707, 1968. 

8. Tidd D, Paterson A. A biochemical mechanism for 
delayed cytotoxic reaction of 6-mercaptopurine. 
Cancer Res 34:738-746, 1974. 

9. Heller E. Enhancement of Chikungunya virus 



replication and inhibition of interferon production 
by actinomycin D. Virology 21:652-656, 1%3. 

10. Wagner RR. The interferons: Cellular inhibitors of 
viral infection. Annu Rev Microbiol 17:285-297, 
1%3. 

11. Wagner RR, Huang A. Reversible inhibition of 
interferon synthesis by puromycin. Evidence for 
an interferon specific messenger RNA. Proc Nat 
Acad Sci USA 54:1112-1115, 1%5. 

12. Calabresi P, Parks R. Chemotherapy of neoplastic 
diseases. In: Oilman A, Goodman L, Oilman L, 
eds. The Pharmacological Basis of Therapeutics. 
New Yoric, MacMiUan, 1980. 

13. Kilboume E, Smart K, Pokomey B. Inhibition by 
cortisone of the synthesis and action of interferon. 
Nature (London) 190:650-651, 1961. 

14. Postic B, DeAngeles C, Brcinig M. Hoe M. Ef- 
fects of Cortisol and adrenalectomy on induction 
of interferon by endotoxin. Proc Soc Exp Biol 
Med 125:89-92, 1%7. 

15. Talas M, Stoger I. Hormone pre treatment and in- 
terferon production in mice. Acta Virol 16:211- 
216, 1972. 

16. DeSomer P, Billeau A, DeClercq E. Influence of 
N'-methylacetamide on interferon production in 
vivo in rats. In: Rita O, ed. Interferons. New 
York. Academic Press, pp65-81, 1968. 

17. Mendelson J, Olasgow L. The in vitro and in vivo 
effects of Cortisol on interferon production and 
action. J Immunol 96:345-362, 1966. 

18. DeMaeyer E, E)eMaeyer J. Two sided effect of 
steroids on interferon in tissue culture. Nature 
(London) 197:724-725, 1%3. 

19. Adolf O, Swetly P. Olucocorticoid hormones in- 
hibit DNA synthesis and enhance interferon pro- 
duction in human lymphoid cell line. Nature 
(London) 282:736-738, 1979. 

20. Havell EA, Vilcek J. Inhibition of interferon se- 
cretion by vinglastine. J Cell Biol 64:716-719, 
1975. 

21. St Oeme J, Horrigan D, Toyama P. The effect of 
6-mercaptopurine on the synthesis and action of 
interferon. Proc Soc Exp Biol Med 130:852-855, 
1960. 



Received March 26, 1982. P.S.E.B.M. 1982, Vol. 171. 



nOCtBDi%€» OF THE SOCIETY FOB EXPERIMENTAL BIOLOGY AND MEDICINE 171, 96-106 (1962) 



Immune Response to Laminin, a Noncollagenous Glycoprotein of Basement 
Membrane, in a Syngeneic Murine System^ (41484) 

ANNE M. MACKEL, FRANK DeLUSTRO,* and E. CARWILE LeROY 

liepartment of Basic and Clinical immunology and Microbiology, and the Division of Rheumatology and 

Immunology, Department of Medicine, Medical University of South Carolina, 

Charleston, South Carolina 29425 



Abstract. The immune response to the noncollagenous, attachment glycoprotein of 
basement membrane (laminin) was studied in a syngeneic murine system. Delayed-type 
hypersensitivity (DTH) was assayed in C57BL/6 mice by measuring footpad swelling fol- 
lowing challenge with connective tissue antigens. Mice receiving a single sensitizing injec- 
tion of laminin in Freund's complete adjuvant (FCA) developed a significant DTH response 
which peaked on Day 7, 24 hr after challenge with laminin or collagenase-treated laminin. 
Laminin-sensitized mice failed to show any significant footpad swelling when challenged 
with types 1 or IV collagen, or fibronectin throughout these experiments. Normal mice 
displayed no significant DTH when challenged with these collagenous or noncollagenous 
connective tissue antigens. Adoptive transfer of laminin-sensitized spleen cells into normal 
mice resulted in significant DTH responsiveness to challenge with laminin; depletion of T 
cells from the immune spleens abrogated this response. Twenty-four hours after challenge 
with laminin, histology of the footpad lesions of laminin-sensitized mice revealed a mononu- 
clear cell infiltrate, characteristic of a DTH response. Mice receiving repeated injections of 
laminin in Freund*s incomplete adjuvant (FIA) developed significant antibody responses as 
detected by the enzyme-linked immunosorbent assay (ELISA). Furthermore Pronase, but 
not coUagenase, treatment of laminin destroyed its antigenicity. Laminin immune sera 
showed no reactivity when assayed on fibronectin or collagen types I-V. No cross- 
reactivity was exhibited by murine anti-type IV or anti-type I collagen antisera with laminin. 
These studies demonstrate the ability of isologous laminin to induce antigen-specific cell- 
mediated and humoral immunity in a murine model. 



Basement membranes are extracellular ran sulfate proteoglycan (7, 8) and entac- 

tissue structures which separate paren- tin (9, 10). 

chymal cells from underlying connective Tissues containing a high concentration 
tissue matrices. Basement membranes pro- of basement membrane (i.e., lung and kid- 
vide a supportive framework for epithelial ney) are frequent sites of immunologically 
(1, 2) and endothelial cells, and function as mediated injury; however, until recently, 
a barrier to the passage of macromolecules specific immunity to basement membrane 
(3). Several basement membrane-specific components has been primarily examined 
proteins have been identified: type IV col- through the use of specific antibodies to aid 
lagen (3), laminin (4), or GP-2(5, 6), ahepa- in the characterization of its biochemical 

' This work was supported by grants from the Na- anti-mouse Thy 1.2 antibody; C, guinea pig comple- 

tional Institutes of Health (AM 20571, AM 21554). the ment; DTH, delayed-type hypersensitivity; EHS, 

South Carolina Chapter of the Arthritis Foundation, Engelbroth- Holm/Swarm sarcoma; ELISA, en- 

thc RGK Foundation, and the Charlotte and Sidney zyme-linked immunosorbent assay; FCA. Freund's 

IJfschuItz Foundation. complete adjuvant; FIA, Freund's incomplete adju- 

^ To whom correspondence should be addressed: vant; PBS-Tween-BSA, PBS (pH 7.8) containing 

Division of Rheumatology and Immunology, Depart- 0.059f Tween 20 and 1% bovine serum albumin; 

ment of Medicine. Medical University of South SDS -PAGE, sodium dodecyl sulfate -polyacrylamide 

Carolina, 171 Ashley Avenue, Charleston, S. C. 29425. gel electrophoresis; T cell, thymus-derived lympho- 

' Abbreviations used: ABTS, 2,2'-azino-di-(3-ethyl- cytc. 
ben/.thia/-oline-6-sulfonate); anti-Thy 1.2, monoclonal 

98 
■ 9727/82/090098- 1 1 $01 .00/0 

I (I > IVK2 by Ihe .Socicly for Kxperimental Biology and Medicine. 
/ie»rrvtd 



IMMUNE RESPONSE TO LAMININ 



99 



properties and tissue distribution. We have 
previously demonstrated the ability of 
types IV and I collagen to induce cell- 
mediated and humoral immunity in a 
murine model (11-13). Antibodies to type 
IV collagen have been identified in the sera 
of patients with systemic sclerosis (sclero- 
derma) (14). The levels of these anti- 
collagen autoantibodies correlated directly 
with the extent of interstitial lung disease 
in this patient population. These studies 
implicate type IV collagen as an autoanti- 
gen and immunity to this basement mem- 
brane protein may be involved in initiating 
and/or perpetuating this disease process. 

Laminin, a disulfide-linked glycoprotein 
is present in all basement membranes in 
quantities approximately equal to type IV 
collagen (4). Although the role of laminin in 
the basement membrane has not been com- 
pletely determined, it appears to function in 
cell attachment similar to fibronectin with 
type I collagen; however, laminin is bio- 
chemically and immunologically distinct 
from fibronectin (4, 15). Sakashita et al, 
(16) demonstrated that laminin specifically 
binds to heparin and heparan sulfate. They 
suggest that an interaction of laminin with 
heparan sulfate proteoglycans of the base- 
ment membrane may contribute to the 
structural integrity of basement membranes 
in vivo. In addition, Ekblom^r al, (17) have 
suggested that laminin may be involved in 
the increased cell adhesiveness observed 
during the aggregation phase of kidney 
morphogenesis. In the present study, we 
have examined the ability of laminin, iso- 
lated from the Engelbroth- Holm/Swarm 
(EHS) sarcoma, to induce an immune re- 
sponse in a syngeneic murine system. Our 
data demonstrate cell-mediated immunity 
to laminin, following a single sensitizing 
exposure, and the development of a strong 
antibody response after repeated immuni- 
zations with this basement membrane 
antigen. 

Materials and Methods. Animals, 
C57BU6 female mice (Laboratory Animal 
Medicine, Medical University of South 
Carolina) were used at 6-8 weeks of age. 

Antigens, Laminin was isolated from the 
EHS sarcoma as described by Timpl et al. 



(4). Tumors were homogenized in 3.4 Af 
NaCl (0.05 M Tris-HCl, pH 7.4) at 4° to 
remove soluble extracellular proteins. The 
residue was extracted twice with 0.5 M 
NaCl (0.05 M Tris-HCl, pH 7.4) at 4°. All 
extractions contained the protease inhib- 
itors phenylme thane sulfonyl fluoride (160 
/jtg/ml; Sigma Chemical Co., St. Louis, 
Mo.) and N-ethylmaleimide (1.25 mg/ml; 
Sigma). Type IV collagen was removed 
from the 0.5 M NaCl extract by increasing 
the salt concentration to 1.7 Af and cen- 
trifuging the solution at 10,000 rpm. The 
supernatant fluid containing laminin was 
dialyzed against 2 M urea (0.05 M Tris- 
HCl, pH 8.6) and passed over a diethyl- 
aminoethyl (DEAE) cellulose column 
equilibrated with the same buffer. The 
unbound material containing laminin was 
concentrated by ultrafiltration (Diaflo fil- 
ter XM 100, Amicon Corp., Lexington, 
Mass.). The concentrated sample was 
chromatographed on an agarose A- 1.5m 
column, equilibrated and eluted with 1 M 
CaClz (0.05 M Tris-HCl, pH 7.4). Laminin 
appeared in the void volume and was ly- 
ophilized following dialysis against 0.05% 
acetic acid. 

Collagen types I and IV were prepared as 
described previously (11, 18). Type I colla- 
gen was isolated from murine tail tendon. 
The EHS sarcoma served as the source of 
type IV collagen. Following pepsinization 
of the homogenized tumor tissue, the colla- 
gen was purified by sequential acetic acid 
solubilizations, precipitation with NaCl, 
and DEAE-cellulose chromatography. 
Human fibronectin was obtained from Col- 
laborative Research Inc. (Waltham, Mass.). 

Analysis of antigens, Laminin and col- 
lagen types I and IV were assayed for 
purity by amino acid analysis, sodium do- 
decyl sulfate -polyacrylamide gel electro- 
phoresis (SDS-PAGE), and susceptibility 
to protease-free bacterial collagenase. No 
evidence of other collagen types or non- 
collagenous impurities were observed in 
the collagen preparations as previously de- 
scribed (11). Following reduction with 2- 
mercaptoethanol, the laminin preparation 
exhibited two homogeneous bands on 
SDS- polyacrylamide gels (4%) with mo- 



100 



IMMUNE RESPONSE TO LAMININ 



lecular weights slightly greater than 200,000 
and 400,000. Treatment of laminin with 
protease-free bacterial collagenase prior to 
analysis by SDS-PAGE did not alter the 
gel patterns. Treatment of types I and IV 
coUagens in the same manner eliminates 
their gel bands (11). Thus the laminin prep- 
aration appears free of collagenous con- 
taminants. 

Collagenase and Pronase treatments, 
Laminin was dissolved in O.OOS M CaCU 
(0.05 M Tris-HCl, pH 7.6) at a concentra- 
tion of 2.5 mg/ml. Bacterial collagenase 
(Clostridium histolyticum; Millipore Corp., 
Freehold, N.J.) was purified by the method 
of Peterkofsky and Diegelmann (19) and 
was free of noncollagenase proteolytic ac- 
tivity as we have previously demonstrated 
(11). Purified bacterial collagenase or Pro- 
nase (Calbiochem-Behring Corp., La 
Jolla, Calif.) was added at concentrations of 
20 units/125 /il and 0.2 mg/125 /il. respec- 
tively. The samples were incubated at 37"* 
for 18 hr, and subsequently dialyzed ex- 
tensively against phosphate-buffered sa- 
line (PBS). 

Delayed'type hypersensitivity. Five mi- 
crograms laminin in 0.1 ml of 0.1 M acetic 
acid was emulsified with an equal volume of 
Freund's complete adjuvant (FCA; Gibco, 
Grand Island, N.Y.) and injected subcuta- 
neously (sc) in the abdomen; control mice 
were untreated. DTH was measured by 
footpad swelling in response to antigenic 
challenge as described previously (11). 
Briefly, mice were injected intradermally in 
the plantar surface of the hindfoot with 5 /ig 
antigen in a volume of 0.03 ml. Footpad 
thickness was determined 4 and 24 hr post- 
challenge with a micrometer and compared 
to measurements observed prior to antigen 
injection. The data are expressed as the 
mean percentage footpad swelling ± stan- 
dard error (SE): 

% footpad swelling 



mm after challenge 

- mm before challenge 

mm before challenge 



X 100. 



Passive transfer. Spleens were removed 
from normal or laminin-sensitized mice 7 



days after immunization. The sensitized 
spleen cells were left untreated or were in- 
cubated with monoclonal murine anti-Thy 
1.2 antibody (Anti-Thy 1.2; New England 
Nuclear, Boston, Mass.) plus guinea pig 
complement (C; M.A. Bioproducts» Wal- 
kersville, Md.) as previously described 
(II). Twenty-five million untreated or T- 
cell-depleted laminin-sensitized spleen 
cells, or 25 x 10® normal spleen cells were 
injected intraperitoneally (ip) in 0.5 ml PBS 
into normal mice. Two days after cell 
transfer, recipient mice were challenged in 
the footpad with 5 /ig laminin and footpad 
swelling was assayed 24 hr later. 

Histology, Normal and laminin-sen- 
sitized mice were challenged in the foot- 
pad on Day 6 with 5 /ig laminin. Twenty- 
four hours later, mice were sacrificed 
and their feet fixed in a neutral glu- 
teraldehyde solution, followed by em- 
bedding in glycol methacrylate (Polysci- 
ences. Inc., Warrington, Pa.). Sections 
were stained with hematoxylin and eosin, 
and examined by light microscopy (11). 

Preparation of antisera. Rabbit anti- 
bodies against types I and IV collagen 
were prepared as described previously (14). 
Rabbit anti-laminin serum was obtained 
using the same immunization protocol. A 
New Zealand white rabbit was injected in 
each hind footpad on Days and 14 with 0.5 
mg laminin in FCA. On Days 23 and 33, the 
animal was injected sc with 1 mg laminin in 
Freund's incomplete adjuvant (FIA; 
Gibco). Blood was collected from the ear 
on Days 30 and 40, and by cardiac puncture 
on Day 41. A gamma globulin fraction was 
obtained following ammonium sulfate pre- 
cipitation (14). 

Mice were immunized on Days 0, 14, 21, 
28, and 35 by sc injections of 50 /ig laminin, 
type IV collagen, or type I collagen dis- 
solved in 0.1 A/ acetic acid and emulsified 
with an equal volume of FIA. Immune and 
normal mice were bled from the ophthalmic 
venous plexus on Day 42. The sera from 
each group (at least four mice per group) 
were pooled and stored at -70°. 

Enzyme-linked immunosorbent assay 
(ELISA), Antibodies to laminin were de- 
tected using the ELISA as we have de- 



IMMUNE RESPONSE TO LAMININ 



101 



scribed previously (13, 14). Serial dilutions 
of rabbit antibodies (1 mg/ml) and immune 
mouse sera in PBS (pH 7.8) containing 
0.05% Tween 20 and 1% bovine serum al- 
bumin (PBS-Tween-BSA), were applied 
to microtiter wells (Flow Laboratories, 
McLean, Va.) coated with 1.25 fig laminin 
and incubated for 45 min. After washing, 
100 fil of peroxidase-conjugated goat anti- 
rabbit immunoglobulin (IgG, M, A) anti- 
bodies (1:500, Cappel Laboratories, Coch- 
ranville. Pa.) or peroxidase-conjugated 
goat anti-mouse immunoglobulin (IgG, M, 
A) antibodies (1:250; Cappel) in PBS- 
Tween-BSA were added, the plates in- 
cubated for an additional 45 min, and 
then washed. One hundred microliters of 
the substrate, 0.03% ABTS (2,2'-azino- 
di-[3-ethylbenzthiazoline-6-sulfonate], 
Sigma), was added to each well in 0.1 
M citrate buffer (pH 4.0) with 0.05% 



H2O2, and after incubation for 1 hr, the ab- 
sorbance was read at 414 nm on a Titertek 
Multiskan (Flow). All assays were per- 
formed in duplicate and the results are ex- 
pressed as the mean absorbance value. 

Absorption assays. Repeated absorptions 
of normal and immune mouse sera were 
performed to remove anti-laminin antibody 
activity. A 1:20 dilution of each serum was 
plated as described above. After a 45-min 
incubation, the sera were transferred to 
other laminin-coated wells. This process 
was continued for up to four absorptions, 
and the assay was completed by the addi- 
tion of peroxidase-conjugated immuno- 
globulin and substrate. In order to control 
for the loss of volume resulting in reduction 
of the sera activity on laminin following 
each transfer, mouse anti-type I and anti- 
type IV collagen sera were plated on lami- 
nin. These sera were then transferred to 



• (0 



o E 
o w 



15-- 



10-- 



5-. 



5-- 




7 11 

Day Po8t-8«n8itization 

FlO. 1. DTH response of (A) mice sensitized with 5 fig laminin in FCA and (B) normal mice 
challenged on Days 3, 6, 10, and 13 postsensitization with laminin (■), type I collagen (ii), type IV col- 
lagen (D), and fibronectin (S). DTH was assayed 24 hr postchallenge and the results are expressed as 
mean percentage footpad swelling ± SE. The number of mice assayed is indicated above each bar. 



102 



IMMUNE RESPONSE TO LAMININ 



plates coated with their respective antigen. 
No significant loss of reactivity with the 
collagen antigens was observed by the 
anti^ollagen sera after the four absorptions 
with laminin. 

Results. The ability of €376176 mice to 
mount a DTH response to homologous 
laminin is shown in Fig. 1. Groups of mice 
were immunized with S fig laminin in FCA 
on Day 0; normal controls were untreated. 
Mice were challenged with S fig laminin, 
type I collagen, type IV collagen, or fi- 
bronectin on Days 3, 6, 10, and 13 postsen- 
sitization. Footpad swelling was assayed 4 
and 24 hr later, with maximal swelling oc- 
curring at 24 hr. No significant swelling 
was observed at 4 hr, indicating that the 
response was not due to an antibody- 
mediated Arthus reaction or to an immedi- 
ate hypersensitivity response. As shown in 
Fig. lA, mice sensitized and challenged 
with laminin demonstrated significant DTH 
on Day 4 (7.2% ± 0.5), with the response 
peaking 7 days postsensitization (13.8% ± 
1.9). Laminin-sensitized mice showed no 
significant DTH when challenged with 
types 1 or IV collagen, or fibronectin. 
Normal control mice (Fig. IB) failed to 
show significant swelling at any time when 
challenged with laminin, types 1 or IV 
collagen, or fibronectin. 

To eliminate the possibility that a col- 
lagenous protein contaminant in our lami- 
nin preparation was contributing to the 
observed DTH response, mice were chal- 
lenged with collagenase and Pronase- 
trcated laminin (Table I). Normal and 
laminin-sensitized mice showed no signifi- 
cant footpad swelling on Day 7, 24 hr after 
challenge with Pronase-treated laminin 
(Table I, Kxpt 1). Challenge of these same 
mice with untreated laminin immediately 
following assay of the footpad swelling on 
Day 7 resulted in significant footpad swell- 
ing (P ' 0.025) on Day 8 in the laminin- 
sensitized mice only. Laminin-sensitized 
mice challenged on Day 6 with collage- 
nasc-lreated laminin displayed significant 
DIM 24 hr later; normal controls showed 
no significant swelling (Table 1, Expt 2). 
Ihcsc data indicate that noncollagenous 
protein antigens are responsible for the 






c 

I' 



c c 
*£ 'c 
EE 



:= 5; . r^. ^ — o 






i^\^9^ :j 



.2 E 



c c 

' E E -3 

. 1j l; u 

Hi 

7 -^ ? .S 

I S r:^ I 

^ ^ o — 



E I «^. "^ f*-. 



II 

O — 



5 I 'E c 

•- o •- o •- 

•7 5 E 5 E 

g ■ z - z - 



|| 



— ri 



^ E 



^1 
u 

£i 

III 

sz g. 

" E & 

a .s .s 

B M 

38 ^ C 
^ fc. > 

^^ ^ 

j: « o 

^^ >. 

— c Q 
.S= c 
.S ^ o 
E S «i 

■5£f 

-I ^ 

« .5 c 

* I* 
■|"sl 

^ 'S3 c 



IMMUNE RESPONSE TO LAMININ 



103 



observed DTH response in the laminin- 
sensitized mice. 

The ability of laminin-sensitized spleen 
cells to adoptively transfer DTH respon- 
siveness to normal syngeneic mice is shown 
in Table II. Spleen cells were obtained from 
laminin-sensitized mice on Day 7, the peak 
of the DTH response (Fig. 1). Twenty-five 
million sensitized spleen cells were injected 
ip into normal syngeneic mice, which were 
challenged 48 hr later with laminin; these 
mice displayed significant footpad swelling 
24 hr after challenge. Mice receiving an ip 
injection of 2S x 10^ sensitized spleen cells 
which were pretreated with a monoclonal 
murine anti-Thy 1.2 antibody plus C failed 
to display significant footpad swelling when 
challenged with laminin. Injection of 2S x 
10* normal €578176 spleen cells into nor- 
mal mice did not transfer reactivity to lami- 
nin. These data indicate that Thy 1.2- 
positive T lymphocytes are responsible for 
mediating the DTH response to laminin. 

Histological examination of the footpad 
lesions of laminin-sensitized mice on Day 7, 
24 hr after challenge with laminin, revealed 
an inflammatory response with a predomi- 
nantly mononuclear cell infiltrate (Figs. 2 A 
and B). Histologic sections obtained from 
the footpads of normal mice after challenge 
with laminin showed no significant inflam- 
mation (Figs. 2C and D). The identification 
of a mononuclear cell infiltrate in the foot- 
pad lesions of the laminin-immune mice is a 
feature characteristic of a DTH response. 

The ability of €578176 mice to elicit an 
antibody response to homologous laminin 
was examined using the ELISA. Rabbit an- 
tisera against laminin were included in all 
assays as positive controls and to determine 
reproducibility of the procedure. Rabbit 



antisera to types I or IV collagen display no 
significant reactivity when assayed on 
laminin-coated wells. No antibody was de- 
tected in the sera of C57BL/6 mice dis- 
playing DTH as a result of a single injection 
of laminin; however, sera obtained from 
mice receiving repeated sc injections of 50 
fig laminin in FIA displayed significant 
antibody reactivity (laminin titers >320) 
when assayed on laminin (Fig. 3). Minimal 
background absorbance was exhibited by 
normal mouse sera at all dilutions and rep- 
resent background values (Fig. 3). Sera 
obtained from mice sensitized to type I and 
type IV collagen using the same immuniza- 
tion schedule as with laminin displayed no 
significant reactivity when assayed on 
laminin (Fig. 3); however, these sera do ex- 
hibit high antibody responses when assayed 
on wells coated with their respective im- 
munogens (13). The anti-laminin mouse 
sera did not react with wells coated with 
fibronectin or types I- V collagen (data not 
shown). 

Treatment of the laminin-coated wells 
with purified bacterial collagenase (11) had 
no effect on the ability of the anti-laminin 
sera to react with laminin (Fig. 4A); Pro- 
nase treatment of laminin-coated wells 
eliminated the observed anti-laminin reac- 
tivity of the immune sera. Repeated ab- 
sorptions of the anti-laminin sera on 
laminin-coated wells were performed to 
determine if all laminin-specific activity 
could be removed, since a single absorption 
did not remove all activity (data not 
shown). As shown in Fig. 4B, reactivity of 
laminin immune sera was reduced to 
background by sequential absorptions on 
laminin-coated wells. Thus, the antibody 
reactivities of these immune sera are spe- 



TABLE II. Adoptive Transfer of 


• DTH TO Laminin Using Untreated and T-Cell-Depleted 
Sensitized Spleen Cells 


Sensitization" 


Cell 
treatment 






Challenge 


No. of Footpad swelling 
mice (mean % ± SE) 


None 

Laminin 

Laminin 


None 

None 

Anti-Thy 1.2 -h C 




Laminin 
Laminin 
Laminin 


4 1.5 ±0.38 
6 9.7 ±1.20 
8 1.3 ±0.36 



Donor mice were either untreated or immunized with 5 /Lig laminin in FCA. 



104 



IMMUNE RESPONSE TO LAMININ 








J^^- ^-^T^ 




Fig. 2. Histological examination of the footpad lesions from laminin-sensitized and normal mice 7 
days postsensitization and 24 hr after challenge with laminin. Sections from a laminin-sensitized mouse 
showing (A) an inflammatory response (x 145) and (B) the presence of a mononuclear cell infiltrate 
(x285). Sections from a normal mouse challenged with laminin showing no inflammation (C, xl45; 
D, x285). 



cific for a noncollagenous protein of the 
basement membrane. 

Discussion. Noncollagenous macro- 
molecules are integral components of base- 
ment membranes interacting with type IV 
collagen to generate this extracellular 
matrix. Laminin, the major noncollagenous 
glycoprotein of basement membrane, has 
been shown by indirect immunofluores- 
cence to be associated with a wide variety 
of mammalian tissues (4, 15) and, by elec- 
tron microscopy, to be localized primarily 
in the lamina rara of the basement mem- 
brane (15, 20). Laminin, consisting of two 
polypeptide chains (220,000 and 440,000 
daltons) linked by disulfide bonds, is bio- 
chemically and antigenically distinct from 



type IV collagen and fibronectin (4). The 
Engelbroth- Holm/Swarm (EHS) sarcoma 
is a transplantable murine tumor which 
produces an extracellular matrix of base- 
ment membrane components (21). The 
present studies were performed using the 
EHS tumor as the source of laminin and 
type IV collagen. Previous investigators 
have demonstrated that antigens expressed 
on type IV collagen and laminin isolated 
from the EHS tumor matrix cross-react 
with normal human and murine basement 
membranes (4, 15, 22, 23). C57BL/6 mice 
were used in the present studies for both 
passage of the EHS tumor line and as hosts 
for laminin sensitization. 
The basis for cell-mediated immunity 



IMMUNE RESPONSE TO LAMININ 



105 




1:20 



1:40 



i:80 



1:160 



r.320 



1:640 



S«rum Dilution 
Fig. 3. Analyses of immune and normal mouse sera on laminin-coated wells using the ELISA. 
Serial dUutions of mouse anti-Iaminin sera (•), mouse anti-type I collagen sera (A), mouse anti-type IV 

collagen sera (■), and normal mouse sera ( ) were assayed on microliter wells coated with 1.25 

/ng lamlnin. The results are expressed as the mean absorbance value at 414 nm. 



(CMI) to connective tissue components was 
established earlier by Adelmann and co- 
workers (24, 2S) and by Senyk and Michaeli 
(26). These investigators demonstrated that 
type I collagen could serve as an antigen for 
the induction of a delayed-type hypersen- 
sitivity (DTH) response; however, they re- 
ported conflicting results regarding the 
ability of an animal to mount an immune 
response to homologous collagens. Senyk 
and Michaeli (26) demonstrated DTH skin 
reactions in guinea pigs to homologous type 
I collagen. However, Adelmann et al. (24, 
25) were unable to demonstrate CMI in 
guinea pigs to homologous type I collagen 
employing similar techniques. These ob- 
servations may be attributable to the strain 
differences of experimental animals. 

Our data demonstrate the ability of ho- 
mologous laminin to induce cell-mediated 
and humoral immunity in a murine system. 
Mice receiving a single sc injection of S /xg 
laminin in FCA developed significant foot- 
pad swelling 24 hr after challenge with 



laminin. This DTH response reached a 
maximal level on Day 7 postsensitization. 
The response was specific for laminin and 
was not observed after challenge with type 
I collagen, type IV collagen, or fibronectin. 
Histological examination of the footpad le- 
sion revealed an infiltration of mononuclear 
cells, indicative of a classical DTH re- 
sponse. Laminin- sensitized spleen cells 
adoptively transferred DTH responsiveness 
to normal syngeneic mice and depletion of 
T cells by prior treatment of the immune 
spleen cells with anti-Thy 1 .2 serum plus C 
eliminated the ability to transfer this re- 
sponse. Thus, mice sensitized to laminin 
mount an antigen-specific DTH response to 
this basement membrane antigen. 

Although antibody was not detected in 
mice displaying DTH to laminin after a 
single injection, higher concentrations of 
the sensitizing antigen and use of protocols 
employing repeated immunization elicited 
significant antibody titers to laminin. No 
cross-reactivity was exhibited by antisera 



106 



IMMUNE RESPONSE TO LAMINIS 



1.0-' 



0.9 -> 



O 



iD 
< 



A 

I I 
_llJ 



EIb 




Number of Abtorptiont 

f'K/. 4. (A) Reactivity of normal (S) and anti-laminin (iO) mouse sera on microtiter wells coated with 
1.25 /Lig (U) untreated laminin. (C) collagenase-treated laminin, and (P) Pronase-treated laminin prep- 
arationn. (B) Repeated absorptions of anti-laminin (O) and normal (■) mouse sera on laminin-coated 
wells. All sera were assayed at a 1:20 dilution on wells coated with 1.25 /Lig laminin. The results are 
expressed us indicated in Fig. 3. 



to types I or IV collagen assayed with lami- 
nin, nor did laminin immune sera react with 
other connective tissue components. Col- 
lagcnase treatment of laminin prior to 
analysis in the HIJSA showed no alteration 
in reactivity of the sera with laminin. 

Although the potential of basement 
membrane to act as an autoantigen has been 
shown, the precise antigens involved are in 
most instances unknown. Cell-mediated 
and/or humoral immunity to basement 
membranes could be directed at either col- 
lagenous or noncollagenous components, 
or both. Data from our laboratory support 
the hypothesis that basement membrane 
damage could be initiated or perpetuated by 
immunity to basement membrane collagen 



(13, 14). Significant titers of anti-type IV 
collagen antibodies were detected in the 
sera of some patients with scleroderma. 
Antibody levels correlated inversely with 
pulmonary diffusion capacity, a reliable 
early indicator of pulmonary interstitial dis- 
ease. These anti-collagen autoantibodies 
were primarily of the IgM class, an immu- 
noglobulin that is a potent activator of the 
complement cascade (13). Gay et al. (27) 
have reported autoantibodies to a collage- 
nous component of basement membrane (C 
chain) in the sera of patients with epider- 
molysis bullosa simplex, a severe blistering 
skin disease. We have previously reported 
the induction of cell-mediated and humoral 
immunity to isologous type IV collagen in a 



IMMUNE RESPONSE TO LAMININ 



107 



murine model (11, 13). Foidart et al, (28) 
demonstrated anti-laminin and anti-type IV 
procollagen antibodies in the sera of pa- 
tients with Goodpasture's syndrome. 
Szarfman et al, (29) have reported that sera 
from humans with Chagas' disease and 
Rhesus monkeys infected with Trypano- 
soma cruzi contain antibodies which react 
with laminin. Furthermore, injection of 
affinity-purified sheep anti-laminin IgG into 
rabbits results in antibody binding to the 
glomerular basement membrane, with sub- 
sequent alterations in glomerular structure 
and proteinuria (30). These studies indicate 
that immunity to basement membrane 
components may be involved in disease 
processes. 

Changes in vascular basement mem- 
branes occur during the progression of dis- 
eases such as scleroderma, atherosclerosis, 
and diabetes mellitus. Alterations in base- 
ment membrane structure and/or assembly 
may reflect endothelial damage, which may 
result in exposure of the basement mem- 
brane and induction of immunity to previ- 
ously sequestered basement membrane 
components. Our observations and those of 
Foidart et al. (28) and Szarfman et al. (29) 
suggest that laminin, the major noncol- 
lagenous glycoprotein of basement mem- 
branes, is a strong immunogen, capable of 
inducing cell-mediated and humoral im- 
munity. Studies are currently underway to 
investigate the possible immunopathologi- 
cal sequelae which may occur following the 
induction of immunity to homologous lami- 
nin in mice. The nature of the body compo- 
nents which elicit an immune response 
(autoimmunity) in connective tissue dis- 
eases is the subject of continuing investiga- 
tion. The development of an animal model 
based on laminin immunity may lead to a 
better understanding of the inflammatory 
responses which occur in human and ex- 
perimental animal diseases involving base- 
ment membranes. 

The authors wish to acknowledge Ms. Judy 
Anderson for preparation of the manuscript, Mr. 
Robert Saper for technical assistance, and Mr. J. J. 
Vandersteenhoven for assistance with the histological 
procedures and photography. 



1. Terranova VP, Rohrbach DH, Martin GR. Role of 
laminin in the attachment of PAM 212 (epithelial) 
cells to basement membrane collagen. Cell 
22:719-726, 1980. 

2. Hogan B. Laminin and epithelial cell attachment. 
Nature (London) 290:737-738, 1981. 

3. Kefalides NA. Structure and biosynthesis of 
basement membranes. Int Rev Connect Tissue 
Res 6:63- 104, 1973. 

4. Timpl R, Rhode H, Robey PG, Rennard SI, 
Foidart JM, Martin GR. Laminin — A glycoprotein 
from basement membranes. J Biol Chem 
254:9933-9937, 1979, 

5. Chung AE, Freeman IL, Braginski JE. A novel 
extracellular membrane elaborated by a mouse 
embryonal carcinoma-derived cell line. Biochem 
Biophys Res Commun 79:859-868, 1977. 

6. Chung AE, Jaffe R, Freeman IL, Vergnes JP, 
Braginski JE, Carlin B. Properties of a basement 
membrane-related glycoprotein synthesized in 
culture by a mouse embryonal carcinoma-derived 
cell line. CeU 16:277-287, 1979. 

7. Kanwar YS, Farquhar MG. Presence of heparan 
sulfate in the glomerular basement membrane. 
Proc Nat Acad Sci USA 76:1303-1307, 1979. 

8. Hassell J, Robey PG, Barrach HJ, WUczek J, 
Rennard SI, Martin GR. Isolation of a heparan 
sulfate-containing proteoglycan from basement 
membrane. Proc Nat Acad Sci USA 77:4494- 
4498, 1980. 

9. Carlin B, Jaffe R, Freeman IL, Chung AE. Isola- 
tion and characterization of sulfated glycoproteins 
from a basement membrane related extracellular 
matrix. Fed Proc 39:1791, 1980. 

10. Carlin B, Jaffe R, Bender B, Chung AE. Entactin, 
a novel basal lamina-associated sulfated glyco- 
protein. J Biol Chem 256:5209-5214, 1981. 

11. Mackel AM, DeLustro F, LeRoy EC. Cell- 
mediated immunity to homologous basement 
membrane (type IV) collagen in C57BL/6 mice. 
Clin Immunol Immunopathol 21:204-216, 1981. 

12. E)eLustro F, Mackel AM, E)eLustro B, LeRoy 
EC. Human monocyte regulation of connective 
tissue growth. Amer Zool, in press, 1982. 

13. Mackel AM, DeLustro F, E)eLustro B, Fudenberg 
HH, LeRoy EC. Immune response to connective 
tissue components of the basement membrane. 
Connect Tissue Res, in press, 1982. 

14. Mackel AM, DeLustro F, Harper FE, LeRoy EC. 
Antibodies to collagen in scleroderma. Arthritis 
Rheum 25:522-531, 1982. 

15. Foidart JM, Bere EW, Yaar M, Rennard SI, Guil- 
lino M, Martin GR, Katz SI. Distribution and im- 
munoelectron microscopic localization of laminin, 
a noncollagenous basement membrane glycopro- 
tein. Lab Invest 42:336-342, 1980. 



108 



IMMUNE RESPONSE TO LAMININ 



16, Safcawhifa S, EngvaO E, Ruoslahti E. Basement 
membrane glycoprocein laminin binds to heparin. 
FEBS Lett 116:243-246, 1980. 

i7. Ekblom P. Alitalo K, Vaberi A.Tmipl R, Saxen L. 
Induction of a basement m embr an e glycoprotein 
in embryonic kidney: Possible rok of laminin in 
moipbofenesis. Proc Nat Acad Sci USA 77: 
485-489, 1980. 

18. Tmtpl R, Bnickner P, Fietzek P. Characterization 
of pepsin fragmems of basement membrane colla- 
gen obtained from a mouse tumor. Eur J Biochem 
^:255-263, 1979. 

19. Peterfcofsky B, Diegelmann R. Use of a mixture of 
proCeinase-free coflagenases for the specific assay 
of radioactive collagen in the presence of other 
proteins. Biochemistry 10:988-994, 1971. 

20. Madri JA, RoO FJ, Furthmayr H, Foidart JM. Ul- 
trastmctural localization of fibronectin and lami- 
nin in the basemem membrane of the murine kid- 
ney, i CeH Biol 86:682-687, 1980. 

21. Orkin RW, Gehron P. McGoodwin EB, Martin 
GR, Valentine T, Swarm R. A murine tumor pro- 
ducing a matrix of basement membrane. J Exp 
Med 145:204-220, 1977. 

22. TimpI R, Martin GR, Bruckner P, Wick G, Wiede- 
man H. Nature of the collagenous protein in a 
tumor basement membrane. Eur J Biochem 
84:43-32, 1978. 

23. TimpI R, GlanviUe RW, Wick G, Martin GR. Im- 
munochemical study on basement membrane 
(type IV) collagen. Immunology 38:109-116, 
1979. 

24. Adclmann BC, Kirranc JA, Glynn LE. The 



structural basis of ceD-medfaaed 
reactions of collagen. CharacterBtics of < 
ous delayed-hypersensitivity reactions in specifi- 
cally sensitized guinea-pigs. Immunology 23:723- 
738, 1972. 

25. Adehnann BC, Kirrane J. The structural basis of 
ceB-mediated inmunological reactions of colla- 
gen. The species specificity of the cutaneous de- 
layed hypersensitivity reaction. Immunology 
25:123-130, 1973. 

26. Senyk G, Michaeli D. Induction of ceB-mediated 
immunity and tolerance to homologous collagen in 
guinea pigs: Demonstration of antigen-reactive 
cells for a self-antigen. J Inununol 111:1381- 
1388, 1973. 

27. Gay S, Ward WQ, Gay RE, Miller EJ. Autoan- 
tibodies to basement membrane collagen: 
Epidermolysis bullosa simplex versus bullous 
pemphigoid. J Cutan Pathol 7315-317, 1980. 

28. Foidart J, Foidart J, Dubois C, Mahieu P. 
Anticorps diriges contre la laminine et le procol- 
lagene de type IV dans le syndrome de Goodpas- 
ture. Nephrologie 1:57-60, 1980. 

29. Szarfman A, Terranova VP, Rennard SI, Foidart 
JM, Lima MDF, Scheinman JI, Martin GR. 
Antibodies to laminin in Chagas' disease. J Exp 
Med 155:1161-1171, 1982. 

30. Abrahamson DR, Caulfield J P. Anti-laminin- 
induced pathology in the rat glomerulus. Fed Proc 
41:1272, 1982. 



Received March 25, 1982. P.S.E.B.M. 1982, Vol, 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 109-113 (1982) 

Factors Modifying DNA Synthesis by Lung Fibroblasts /n Vitro^ (41485) 
C. WILLIAM CASTORS and TERRENCE D. FREMUTH 

The Rackham Arthritis Research Unit, Department of Internal Medicine, The University of Michigan 
Medical School, Ann Arbor. Michigan 48109 



Abstract. DNA synthesis in guinea pig lung fibroblast cultures was shown to be stimu- 
lated by endotoxins, PGE2, CTAP-III, CTAP-Pj, and insulin; indomethacin and Cortisol 
partially reversed some of these effects. DNA synthesis in human lung fibroblasts was also 
markedly stimulated by CTAP-III, while the response to endotoxins was less striking. 



Lung fibroblast cultures derived from the 
gas exctiange component of pulmonary 
parenchyma provide a potentially useful in 
vitro model system for studying pulmonary 
connective tissue metabolism (1-4). This 
culture system permits one to study new 
synthesis of connective tissue by lung fi- 
broblasts, and to study separately the repli- 
cation of the cells and the chemical anat- 
omy of the extracellular matrix materials 
secreted by pulmonary fibroblasts as these 
various activities may be modified by 
physiologic, pathologic, and pharmacologic 
agents. 

In the present study, we focused on lung 
fibroblast proliferative responsiveness 
(measured by incorporation of [methyl- 
^H]thymidine into fibroblast DNA) to rele- 
vant agonists of both exogenous and en- 
dogenous origin. Our interest in this aspect 
of pulmonary inflammation originates in the 
possibility that an expanding population of 
lung fibroblasts may be an important 
antecedent to accelerated fibrosis and de- 
position of extracellular matrix materials. 
Supporting this idea is evidence from ex- 
perimental cotton pellet granulomas in rats 
showing a direct correlation between the 
amount of connective tissue DNA and the 
accumulating extracellular connective tis- 
sue matrix materials of the progressing in- 
flammatory process (S). The inflammatory 
response in rat lung to intermittent oxygen 
exposure was also accompanied by ele- 



* This study was supported by U.S. Public Health 
Service Grant HL-19685. 
^ To whom correspondence should be addressed. 



vatcd DNA values during the later '*fibro- 
blastic'' phase of inflammation (6). Rabbit 
lung fibroblast cultures exposed to various 
dusts often responded with increased DNA 
synthesis in parallel with increased synthe- 
sis of extracellular matrix materials includ- 
ing collagen and glycosaminoglycans (7). 

If fibroblast proliferation is important to 
mounting a destructive fibrotic process in 
pulmonary tissue, factors capable of ac- 
celerating or repressing lung fibroblast 
DNA synthesis may have practical signifi- 
cance. With this in mind, we have exam- 
ined the effect of gram-negative endotoxins 
analogous to what might be released in lung 
during bacterial pneumonitis. Further, 
since immune complex-mediated injury to 
pulmonary endothelium results in local se- 
cretion of platelet-derived growth factors, 
we have examined the effect of two 
platelet-derived connective tissue activat- 
ing peptides, CTAP-III (8, 9) and CTAP-P2 
(10). In addition, we examined the effects of 
prostaglandin E2 (PGE2), known to be se- 
creted by lung fibroblasts (11). 

Materials and Methods. Culture meth- 
ods, Guinea pig lung fibroblasts were iso- 
lated by enzymatic disaggregation of lung 
parenchyma and plating at high density in 
plastic flasks as previously reported (1). 
Human lung fibroblast cultures were estab- 
lished in the same manner from normal tis- 
sue removed at lobectomy for neoplasm. 
Cultures derived in this way were propa- 
gated using medium F-12 (with 10% fetal 
calf serum, PCS 10%) or medium 1066, 
90%: PCS 10%. Three different guinea pig 
lung fibroblast lines and one human lung 
fibroblast line were used in this study. 



109 
0037-9727/82/090109-05$Ol !m<\ 

Copyright © V¥a \si l>*e VkVkX^ toi ^xver«s««aa. V^««i «A^^^ 
Att righU reserved. 



110 



DNA METABOLISM IN LUNG FIBROBLASTS 



Complete medium changes were carried out 
three times a week and trypsin dispersal 
was performed as required for propagation 
or study. Freezing procedures were as pre- 
viously reported. 

DNA synthesis was measured by deter- 
mining the incorporation of [methyl'^H]- 
thymidine into cells in microtiter well 
cultures utilizing 10^ cells per culture (8). 

Sources of mediators, CTAP-III was 
isolated from outdated human platelets by 
methods previously reported (8, 9). This 
cationic protein appears to be the major 
human platelet derived growth factor; it 
was essentially homogeneous in multiple 
PAGE systems and its sequence has been 
established (12). A minor platelet-derived 
growth factor, CTAP'P2, is an anionic ma- 
terial which stimulates DNA and GAG 
synthesis in human synovial cells. CTAP'P2 
was partially purified as a by-product of the 
isolation of CTAP-III, i.e., from biologi- 
cally active fractions which did not react 
with antisera to CTAP-III (10). The major 
contaminant identified in CTAP-P2 prepa- 
rations is albumin. 

The enteric endotoxins were obtained 
from Difco Laboratory, Detroit, Michigan, 
while the Klebsiella pneumoniae LPS was a 
gift from Dr. A. I. Braude, University of 
California, San Diego. Crystalline bovine 
insulin was obtained from Sigma Chemical 



Company, St. Louis, Missouri, and poly- 
inosiiiic:polycytidylic acid (Poly I:C) was 
a gift from Dr. A. A. TyteU, Ph.D., Merck 
Sharp and Dohme, West Pbint, Pennsyl- 
vania. Prostaglandin E2 (PGEt) was kindly 
provided by Dr. John E. Pike, The Up- 
john Company, Kalamazoo, Michigan. 

Results. Endotoxin stimulation of DNA 
synthesis. Bacterial endotoxin was shown 
to be capable of substantially stimulating 
the synthesis of DNA in lung fibroblast 
cultures (Table I). It is noteworthy that 
very small amounts of LPS were not un- 
commonly more stimulatory than higher 
concentrations. It is clear that among the 
potential inhibitors of DNA synthesis, that 
cycloheximide virtually ablated basal syn- 
thesis, indomethacin at clinically achiev- 
able concentrations was without effect, and 
Cortisol exhibited marginal suppression of 
basal DNA synthesis. 

It was noteworthy that both Cortisol and 
indomethacin partially reversed the stimu- 
latory effect of E, coli LPS on lung fibro- 
blast DNA synthesis. The observations 
recorded for E. co/i;026:B6 endotoxin was 
representative of the guinea pig lung fibro- 
blast response to other endotoxins includ- 
ing: S, typhosa'(m\, E, co//;0111:B4, S. 
typhimurium, and 5. minnesota. 

Of some interest was the observation 
(Tables I and II) that PGE2, 1 /xg/ml, signifi- 



TABLE 1. Agents Modifying Endotoxin Stimulation of pH]DNA Synthesis in 
Guinea Pig Lung Fibroblast Cultures" 





PHJDNA 


Experimental/ 




Additives 


(cpm/10* cells)* 


control 


P 


0.15iV/NaCI 


3,787 ± 584 








PBS'^ 


4,271 ± 708 


— 


— 


E. coli 026:36, LPS, 1.0 fig/m\ 


24,380 ± 5935 


6.4 


<0.01 


E. coli LPS, 50 /Ltg/ml 


14,045 ± 3684 


3.7 


<0.01 


Cortisol, 1.0 /Ltg/ml 


2,781 ±716 


0.7 


<0.05 


Cycloheximide. 10 MS/nil 


110 ± 17 


0.03 


<0.01 


Indomethacin. 15 Mg/ml 


2,962 ± 1077 


0.8 


NS" 


E. coli LPS, 50 Mg + Cortisol 


4,799 ± 1980 


1.3 


<0.01 


E. coli LPS, 50 /ig -♦- indomethacin 


8,227 ± 1562 


2.2 


<0.02 


E. coli LPS, 50 /ig + cycloheximide 


69 ±21 


0.02 


<0.01 


PGE,, I.OMg/ml 


7,032 ± 1859 


1.9 


<0.01 



" Target cultures were GP-22 lung fibroblasts. 

" Data are expressed as the mean ± I SD for four to six microcultures in this and subsequent tables. 

' PBS refers to buffered saline containing 0.15 M NaCI plus 0.05 M phosphate buffer, pH 7.0. 

** Not significant. 



DNA METABOLISM IN LUI4G F1MKMU.ASrS 



III 



TABLE n. Agents Moxhfmmg I^atelet oh Inswlin SmyiirLATioN of f^HPN A S\>.THF«rs 
iM Glinea Pig Lung Pimioblast Cultlties* 





PWPNA 


Experifnemjil 




AdditTMcs 


<cpiii 10« celhi 


o«im>l 


P 


0.15 VNjO 


7,tt6 - I7lg 





_ 


PBS 


6,^1 - 1711 


— 


— 


CTAP-nL 21 MS/nA 


39,123 :^ 2006 


5.0 


<0.01 


CTAP^in ^ inAwDfliiarw 


23 JS3 2: 97 


3.0 


<0.01 


CTAP-m -^ ooftifioil 


35,722 ± 241« 


4.6 


NS* 


CoftisoL LOM/sd 


10,405:^:96$ 


13 


<0.05 


lodooietlucin, 15 figfnA 


7,175 ± 412 


0.9 


NS 


CTAP-P^ 163 M^m] 


15,793 2: 1615 


20 


<0.01 


CTAP-P, + mrtntnctturin 


«,967 - 1792 


1.1 


<0.0l 


CTAP-Pj -^ ooftifiol 


10,«2I i^ 2375 


1.4 


<0.05 


Insulm, 0^ Qoit'inl 


40,450 ± 3516 


5.2 


<0.01 


Insulia. 5.0 unitiii] 


42,258-4311 


5.4 


<t>.Ol 


Insulia. 5 uut/nil -k mdomcfhacin 


28,016 r 973 


3.6 


-^0.05 




47,519 ± 1364 


6.1 


NS 


E. coU LPS. a26:B6, 50 fig/mi 


11,263 ±339 


1.6 


<.0.02 


PGE^l.OMg/ml 


17,705 ± 1376 


2,3 


<;0.01 


PGE„5«/iiiI 


11,545 ±1366 


1.5 


-.0.01 



* The GP-20 hiog fibroblast line served as target cultures in this experiment. 

* Not significant. 



cantly stimulated DNA synthesis. Not all 
experiments demonstrated PGEs stimula- 
tion of DNA synthesis; on occasion, very 
high concentrations (S fig/m\) were inhib- 
itory while 0.1 fig/ml was stimulatory. 

Stimulation of DNA synthesis by platelet 
factors and insulin. It was clear that both 
platelet factors, CTAP-III and CTAP-P,, 
stimulated DNA synthesis in guinea pig 
lung fibroblast cultures to an extent compa- 
rable to that in the endotoxin experiments 
(Table ID. In the present experiment, the 
specific stimulation by CTAP-III seemed to 
be greater than that generated by CTAP-Pj, 
although the less purified state of the 
CTAP-P2 makes such calculations impre- 
cise. Indomethacin significantly reduced 
the stimulatory effect of CTAP-III while 
Cortisol was apparently ineffective. Both 
indomethacin and Cortisol significantly re- 
versed the stimulatory effect of the minor 
platelet factor, CTAP-P2. 

Bovine insulin in pharmacologic concen- 
trations was shown to be a potent stimula- 
tor of DNA synthesis in these cell cultures, 
an effect which was resistant to Cortisol 
suppression. On the other hand, indo- 
methacin was able to partially suppress 



the insulin-stimulated increase in DNA 
synthesis. 

Stimulation of DNA synthesis in human 
lung fibroblast cultures. Lung fibroblast 
cultures of human origin showed a marked 
increase in DNA synthesis on exposure to 
the major platelet mitogen CTAP-UU and 
relatively minor levels of stimulation by 
the synthetic polynucleotide. Poly I:C, 
(Table III). 

Insulin caused a minor increase in DNA 
synthesis, cycloheximide ablated DNA 
synthesis, and Cortisol exhibited a border- 
line stimulatory effect. The enteric endo- 
toxins had no signiflcant effect on DNA 
synthesis; K, pneumoniae LPS caused a sig- 
nificant, but minor, stimulation of lung fi- 
broblast DNA synthesis. POEa (data not 
shown) did not stimulate DNA synthesi!<i in 
the human lung Fibroblast cultures. 

Discussion. In general, endotoxin stimu- 
lation of DNA synthesis was greater in 
guinea pig lung flbroblast cultures than in 
our human line. Earlier studies demon- 
strated that the lipid A core of endotoxins 
was a potent stimulator of giycosaminogly- 
can synthesis in human synovial cultures 
(13), and recently lung fibroblasts were 



112 



DNA METABOLISM IN LUNG FIBROBLASTS 



TABLE III. Stimulation of pH}DNA Synthesis by Human Lung Fibroblasts 



Mediator or 


PHJDNA 


Experimental/ 




vehicle 


(cpm/10* cells) 


control 


P 


Experiment 1 








0.15A/NaCl 


1,074 ±105 


— 


^~ 


CTAP-UI, 25 Mg/ml 


11,069 ±4532 


10.3 


<0.01 


Poly I:C. 50 ptg/ml 


2,243 ± 414 


2.1 


<0.01 


Experiment 2 








0.15A/NaCl 


11,417 ±1060 


— 





CTAP-III, 19 fig/ml 


35,636 ± 2076 


3.12 


<0.01 


Insulin, 0.4 fig/ml'' 


16,669 ± 1634 


1.46 


<0.01 


Cortisol, \.Ofig/m\ 


14,181 ±1197 


1.24 


<0.05 


Cycloheximide, lOpig/ml 


1,944 ±273 


0.17 


<0.01 


E. coli 026: B6 endotoxin 


12,962 ± 1736 


1.14 


NS 


0.5 fig/ml 








S. tymphimurium endotoxin. 


12,943 ± 1579 


1.22 


NS 


5.0 Mg/ml 








Klebsiella endotoxin. 


14,240 ± 733 


1.25 


<0.01 


0.5 Mg/ml 









' 0.4 11% corresponds to 10 munit. Normal human insulin values range from 4 to 10 /xunit/ml of serum. 



shown to exhibit increased GAG synthesis 
on exposure to endotoxins (14). Whether 
this relative resistance of our human lung 
fibroblast line to endotoxin stimulation of 
DNA synthesis is a peculiarity of the cell 
line or representative of such cells remains 
to be established. 

These data demonstrate that CTAP-III, a 
growth factor derived from human plate- 
lets, markedly stimulates DNA synthesis 
in lung fibroblast cultures of guinea pig 
and human origin. Since CTAP-III is 
secreted by platelets in the course of the 
aggregation-granule release process (8, 15), 
it is likely that relatively large concentra- 
tions of this mediator flood local areas of 
injury. While it is possible that CTAP-P2 is 
also presented to injured tissue in the same 
manner, there is yet no unambiguous data 
on its secretion mechanism. One may 
speculate that an injurious process in lung, 
be it physical, chemical, immunologic, or 
microbiologic, may initiate fibroblastic 
proliferation in part through this mecha- 
nism. In guinea pig lung cultures, it was 
possible to partly suppress this response 
with the anti-inflammatory agent indo- 
methacin in '^clinically relevant'' con- 
centrations. 



1. Castor CW, Heiss PR, Gray RH, Seidman JC. 
Connective tissue formation by lung cells in vitro. 
Amer Rev Resp Dis 120:107-119. 1979. 

2. Castor CW, Wilson SM, Heiss PR. Seidman JC. 
Activation of lung connective tissue cells in vitro. 
Amer Rev Resp Dis 120:101-106, 1979. 

3. Seidman JC, Castor CW. Connective tissue acti- 
vation in guinea pig lung fibroblast cultures: Reg- 
ulatory effects of glucocorticoids. In Vitro 
17:133-138, 1981. 

4. Castor CW. Fremuth TD. Lung fibroblast metab- 
olism: Factors regulating glycosaminoglycan 
synthesis. Clin Res 28:752A, 1980. 

5. Castor CW. Connective tissue activation. V. The 
flux of connective tissue activating peptide during 
acute inflammation. J Lab Clin Med 81:95-104, 
1973. 

6. Valimaki M. Juva K. Rantanen J, Ekfors T, 
Niinikoski J. Collagen metabolism in rat lungs 
during chronic intermittent exposure to oxygen. 
Aviat Space Environ Med 46:684-690, 1975. 

7. Hext PM. Richards RJ. Biochemical effects of as- 
bestiform minerals on lung fibroblast cultures. 
Brit J Exp Pathol 57:281-285, 1976. 

8. Castor CW, Ritchie JC, Scott ME, Whitney SL. 
Connective tissue activation. XI. Stimulation of 
glycosaminoglycan and DNA formation by a 
platelet factor. Arthritis Rheum 20:859-868, 
1977. 

9. Castor CW, Ritchie JC, Williams CH, Scott ME. 
Whitney SL. Myers SL, Sloan TB, Anderson B. 



DNA METABOLISM IN LUNG FIBROBLASTS 



113 



Connective tissue activation. XIV. Composition 
and actions of a human platelet autacoid mediator. 
Arthritis Rheum 22:260-272, 1979. 

10. Castor CW, Cobcl-Geard SR. Connective tissue 
activation: Evidence for a second human platelet 
growth factor. Clin Res 28:139A, 1980. 

11. Bryant RW, Feinmark SJ, Makheja AN, Bailey 
JM. Lipid metabolism in cultured cells. Syn- 
thesis of vasoactive thromboxane Aj from [**C] 
arachidonic acid by cultured lung fibroblasts. J 
Biol Chem 253:8134-8142, 1978. 

12. Walz DA, Castor CW. Connective tissue activa- 
tion: Structural studies on a human platelet mito- 
gen. Clin Res 27:649A, 1979. 

13. Buckingham RB, Castor CW. The effect of bacte- 



rial products on synovial fibroblast function: 
Hypermetabolic changes induced by endotoxin. J 
CUn Invest 51:1186-1194, 1972. 

14. Castor CW, Fremuth TD, Furlong AM. Lung fi- 
broblast metabolism: Characterization of pro- 
teoglycan (PC) and glycosaminoglycan (GAG) 
synthesis. Clin Res 29:738A, 1981. 

15. Castor CW, Cobel-Geard SR, Hossler PA, Kelch 
RP. Connective tissue activating peptide-III. 
XXII. A platelet growth factor in human growth 
hormone deficient patients. J Clin Endocrinol 
Metab 52:128- 132, 1981. 



Received April 16, 1982. P.S.E.B.M. 1982, Vol. 171. 



nU^.f^JMHijk Of IHt UJCitlY FCm tXrtUMtSlAL BIOLOGY AS'D MEDfOSE 171« 114-119(19821 



Spontaneous Production of High Levels of Leukocyte (a) Interferon by a Human 
Lymphoblastoid B-Ceil Une(LDV/7) (41486) 

a V, ABLASHI,* ' S, BARON,+ G. ARMSTRONG,* A. FAGGIONI/-* D. VIZA,t 

P. H. LEVINE,* AND G. PIZZA§ 

"Naiionat Canter Intlilute, Bethesda, Maryland 20205: ^Department of Micnfbiology. University of Texas, 

Oalvet/on, Texas; XINSERM U.2I2, 29 rue Manin, 75019 Pans, France: 

Wspedale M. Malpighi, Bologna, Italy 



Ahttract. A humsai lymphoblastoid B-cell line (LDV/7) capable of replicating significant 
IcveU aftrdn%fer factor and immune RNA was found to produce high quantities of interferon 
continuously. The maximum yield of interferon (600 units/ml) was detected on the seventh 
dity. The interferon wa« characterized as human leukocyte (a) type. This interferon exhib- 
ited broad antiviral activity against vesicular stomatitis virus, poliovirus, sindbis virus, 
Hfrpr%viru% %aimiri, and Epstein -Barr virus. Treatment of LDV/7 cells with TPA. PHA, or 
l^PS failed to enhance interferon production. Furthermore, LDV/7 cells could not be in- 
duced with Sendai virus to produce larger amounts of interferon. Interferon production 
correlated best with cell proliferation. It is therefore possible that this cell line may be a good 
Miurce for large-scale production of human leukocyte interferon as well as for cloning of the 
interferon genome, and for studies of gene expression. 



A human lymphoblastoid cell line (LDV/7), 
CHtablinhed in 1968 from the peripheral 
blood leukocytes of a healthy 6S-year- 
old male, has been shown to be the only 
lymphoblastoid cell line capable of rep- 
licatinK significant quantities of transfer 
(actor ( no and immune RNA (K 2). Other 
human lymphoblastoid cell lines (B or T), 
or cell lines established from Burkitt's 
lymphoma (.1, 4), including Namalwa (3), 
were found to be unsatisfactory for replica- 
tion of ri". Since LDV/7 is a B-cell line, 
possessing a unique ability to replicate TF, 
wc wtMc interested: (a) in investigating its 
ability to produce interferon spontaneously 
or by induction with Sendai virus or other 
viruses, as has been reported previously for 
other lymphoblastoid cells (5-8), and (b) in 
testing the antiviral activity of this inter- 
Ifion, particularly on two primate on- 
cogenic herpesviruses, Epstein -Barr virus 
(I'HV) of man and Herpesvirus saimiri 
(IIVS) of the squirrel monkey. 

Materials and Methods. The charac- 
teii/ation of I.I)V/7 cells, including sur- 

' 1 n w hi)in nil correspondence should be addressed. 
' iVinianenl address: Istituto di Patologia Gencrale, 
II CaUedia UMivcrsita' di Roma. Italy. 



face markers, cytochemistry, immuno- 
globulin secretion, HLA type, chromosome 
analysis, presence of EBV DNA and 
antigens, and induction of tumors in 
athymic nude mice has been reported 
elsewhere (9). LDV/7 cells were grown at 
37° in RPMI-1640 medium (Gibco) con- 
taining 10% heat-inactivated fetal bovine 
serum and antibiotics. Unlike some B-cell 
lines, LDV/7 cells are unique in their 
growth pattern since more than 15% of the 
cells firmly attach to the plastic surface, 
and clumps of cells, both attached and 
floating, are consistently present in culture. 

Treatment of cells for interferon produc- 
tion. LDV/7 cells were seeded at 2 x lO* 
cells/ml in RPMI 1640 medium containing 
10% fetal calf serum. At the time of seeding 
>95% cells were viable. These cells were 
divided into various groups, as described 
below, for treatment with the tumor pro- 
moter 12-0-tetradecanoylphorbol-13-acetate 
(TPA) and Sendai virus. Supematants were 
harvested at various times and were 
clarified at 300^', filtered through a 0.45-/Ltm 
filter and frozen in liquid nitrogen. The pro- 
cedure was as follows: 

1 . LDV/7 cells supernatants harvested at 
3, 7, 10, and 14 days: 



114 

'h:()*x)ii4(k>$()i.(hvo 

Mil* h» r HxperimcntiiJ Biology and Medicine. 



SPONTANEOUS INTERFERON PRODUCTION 



115 



2. LDV/7 cells treated with 20 ng/ml of 
TPA for 3, 7, and 14 days and supernatants 
harvested as for untreated LDV/7 cells; 

3. LDV/7 cells infected with 4 HA unit of 
Sendai virus and supernatants harvested as 
for TPA-treated cells; 

4. LDV/7 cells treated with phytohemag- 
glutinin (PHA-P), 2 /Ltg/ml and Escherichia 
coli lipopoly saccharides (LPS), 25 /LAg/ml; and 

5. LDV/7 cells treated first with PHA-P 
(2 /Ltg/ml) and then with 20 ng/ml of TPA. 

Assays for interferon. The assays were 
performed by the VSV plaque reduction 
method as described previously (10, 11). 
The titers are expressed as international 
reference units per milliliter. 

EBV and HVS sensitivity to LDVI7- 
produced interferon. This was carried out 
according to the methods previously de- 
scribed (12). The Ag-876 strain of EBV 
from cells grown at 34° was used for induc- 
tion of EBV capsid antigens (VCA) and 
transformation of human cord blood lym- 
phocytes (HCBL) (13). The titer of this 
virus had previously been shown to be \(fi 
transforming units/ml in HCBL. 

Results. Growth characteristics of 
LDV/7 cells, LDV/7 cells form clumps con- 
sisting of 5-400 cells. These clumps either 
attach to the plastic surface or float in the 
culture medium. On subculture, the floating 
cells and/or attached cells were found to 
attach to the plastic surface as well as float 
in the medium. LDV/7 cells replicate faster 
than Raji, Ag-876, P3HR-1 cells, and 
Namalwa (6). 

Interferon production. The spontaneous 
levels of inhibitor detected in the superna- 
tants of LDV/7 cells at various times are 
presented in Fig. 1 . The peak of inhibitor 
production (approximately 600 units) was 
on the seventh day and thereafter the level 
of interferon declined. 

The LDV/7 cells treated with TPA 
showed only slight enhancement of inhib- 
itor production by 0.2io (Fig. 1) on Day 7; 
however, on Day 3 approximately 500 units 
of interferon were detected in TPA-treated 
cells (an increase of 380 units). On Day 14, 
TPA-treated samples showed an increase of 
only 30 units over the spontaneously pro- 
duced inhibitor. 



O 3 



a 
o 

cc 



O 

cc 

it! 

cc 




10 



14 



DAYS SUPERNATANT 
FROM LDV/7 CELLS WERE COLLECTED 

Fig. 1. Detection of interferon (log) in LDV/7 lym- 
phoblastoid cell cultures in the presence and absence 
of TPA and Sendai virus. O, LDV/7 supernatant only; 
•, LDV/7 supernatant with the presence of TPA; A, 
LDV/7 supernatant harvested after infection with Sen- 
dai virus. 

The Sendai virus infected LDV/7 cells 
showed a significant drop in the level of the 
inhibitor (Fig. 1) indicating that LDV/7 cells 
contain a considerable level of inhibitor 
used to neutralize the virus. 

LDV/7 cells treated with 2 /Ltg/ml of 
PHA-P or 25 /Ltg/ml of LPS for 3 and 7 days 
failed to induce any increase in the level of 
spontaneously produced inhibitor (^120 
units Day 3 and ^400 units Day 7). How- 
ever, the treatment of LDV/7 cells first with 
PHA and then with TPA resulted in pro- 
ducing ^600 units of the inhibitor, but this 
effect decreased significantly by Day 7 
(=^160 units). 

Characterization of the inhibitor pro- 
duced by LDV/7 cells. As presented in 
Table I, the inhibitor was characterized as 
being interferon. The inhibitor showed the 
following major properties of human leuko- 
cyte interferon: (a) the inhibitor was active 
against different viruses: sindbis virus, 
pohovirus, and vesicular stomatitis virus; 
(b) the inhibitor was species specific in that 
it protected human WISH and human 
Hep-2 cells but it did not protect mouse L 
cells against virus infection; (c) the inhibitor 
was stable to acid treatment (pH 2) at 4° for 
24 hr; (d) the antiviral activity was cell 
mediated since the washing of the cells 
treated for 24 hr with the inhibitor did not 
remove the resistance to virus infection; (e) 
the inhibitor was inactivated by antiserum 



116 



SPONTANEOUS INTERFERON PRODUCTION 



TABLE I. Characterization of Virus Inhibitor" 















pH2 






Human 


Human 


Human 


Mouse 


Human 


Sample 


Cell type: 


WISH 


WISH 


Hep-2 


L 


WISH 


treatment 
















Virus: 


VS 


Sindbis 


PoUo 


VS 


VS 


LDV/7 supernatant 














harvested Day 3 




120 


>100 


>100 


— 


100 


LDV/7 supernatant 














harvested Day 7 




600 


>100 


>100 


<3 


100 


LDV/7 cells treated 














with20ng/mlofTPA 














supernatant, harvested 














Day 3 




500 


>100 


>100 


— 


— 


LDV/7 cells treated 














with20ng/mlofTPA 














supernatant, harvested 














Day 7 




1000 


>100 


>100 


<3 


100 


LDV/7 ceUs infected 














with 4 HA units of 














Sendai virus, supernatant 














harvested Day 7 




500 


>100 


>100 


— 


— 



<* Data are expressed in international reference units/ml. 

Inhibitor characteristics: (1) Broad antiviral activity; (2) cell species specificity; (3) stable at pH 2; (4) the in- 
hibitor is cell associated since the inhibitor is not removed after three washings of the cells before vims chaUenge; 
(5) the inhibitor was destroyed by trypsin and not by DNase or RNase (data not shown); (6) the inhibitor was not 
sedimented by ultracentrifugation. 



directed against leukocyte interferon (Table 
II); and (f) the inhibitor was destroyed by 
trypsin but not by DNase or RNase and was 
not sedimented by ultracentrifugation. 

Antiviral activity. Results of LDV/7 lym- 
phoid cell interferon on oncogenic HVS of 
nonhuman primates and EB V are presented 
in Table III. Continuous post-treatment of 
owl monkey kidney (OMK) cells with 
LDV/7 supernatant harvested Day 7 re- 
duced the HVS titers by 3 logs. If the OMK 
cells were pretreated with the interferon, a 
further half-log reduction in HVS titer was 
observed. 



The interferon from LDV/7 cells failed to 
induce any effects on intrinsic EB V genome 
expression in Raji cells, since no reduc- 
tion or inhibition of spontaneous expres- 
sion of EBV nuclear antigen (EBNA) was 
observed. This interferon also failed to 
block VCA-EA expression in EB virus pro- 
ducers Ag-876 or P3HR-1 ceUs (Table III). 
Superinfection of Raji cells with lytic 
P3HR-1 EBV normally leads to expression 
of early antigens (EA). Table III shows that 
after superinfection of Raji cells with 10*® 
EA inducing units, approximately 10% EA 
were detected by indirect IF. More than 



TABLE II. 


Antigenic Typing of 


LDV/7 Interferon'* 






Medium 


Antiserum 


I 


Interferon sample 


Anti-fibroblast 
ifi) IFN 


Anti-leukocyte 
(a) IFN 


From LDV/7 cells 
Control fibroblast 03) 
Control leukocyte (a) 
Control immune (y) 


25 

65 

6 

25 


25 

<1 

6 

25 


<10 
65 
<1 

25 



" The interferon titers are expressed after mixing the LDV/7 and control interferons with antiserum to fibro- 
blast and leukocyte interferons. 



SPONTANEOUS INTERFERON PRODUCTION 



117 



TABLE III. Inhibition of Herpesvirus Saimiri-Induced CPE and EBV EA and Transformation in the 
Presence of Supernatant Collected from LDV/7 Cells, after 7 Days of Incubation 

Inhibition 



Cells/treatment/virus 



In the presence of 
LDV/7 supernatant 



In the absence of 
LDV/7 supernatant 



3. 



(a) OMK cells were infected 
with serial dilutions of HVS 
with 10* »/mI TCID50 titer 
for 2 hr, cells were washed 
and then continuously kept at 
37* in maintainance medium with 
and without LDV/7 supernatant 
for 14 days and observed for 
HVS-induced CPE 

(b) Pretreatment of OMK cells 
with LDV/7 spnt. for 24 hr 

at 37* before HVS infection 
Rfyi cells were infected 10*-° 
EA inducing units/ml with P3HR1 
EBV for 90 min. Then the cells 
were washed and fed with LDV/7 
supernatant for 72 hr 
Ag-876 cells growing at 34'' 
were subcultured at seeding 
density of ^1 x lO'/ml and were 
fed with medium with or without 
LDV/7 supernatant for 7 days 
HCBL separated by plasmagel were 
pretreated with LDV/7 supernatant 
(containing 600 units of interferon) 
at 37* for 24 hr. The other set 
of cord cells were kept without 
treatment with LDV/7 supernatant. 
The supernatant was removed and 
cells were washed with medium, 
infected with Ag-876 EBV with a 
titer of \(fVm\ TFU, and observed 
for EBNA and transformation 



10»Vml 


10*Vml 


10» * to 10» «/ml 


10*Vml 


^1.0% cells 


^10% cells 


expressing EA 
and no VCA 


expressing EA 
and 3% VCA 



10-15% EA/VCA 



10-15% EA/VCA 



10»« 



10*« 



90% EA were inhibited when cells were 
kept for 72 hr in supernatant collected from 
7 day-incubated cells. HCBL pretreated for 
24 hr with LDV/7 supernatant containing 
600 U/ml of interferon were infected with 
various dilutions of transforming Ag-876 
EBV. The data in Table III show that up to 
10 transforming units of virus were inhib- 
ited, based on EBNA induction, colonies of 
transformed cells, and outgrowth of these 
cells under soft agar. 

Discussion. The spontaneous production 
of interferon from other human lympho- 
blastoid cell lines has previously been re- 
ported (6, 7, 14). Pickering e/ aL (7) showed 
that with the exception of two B-cell lines 



(8876-7 and Lukll) all other cell lines (Raji, 
CAH, Namalwa, BM, NBRL-AG57) either 
produced small amounts of interferon 
(10-200 units) or lacked interferon produc- 
tion (<10). Similariy Strander et aL (6) 
showed that the majority of the cell lines 
they tested produced <20 units of inter- 
feron; however, cell lines such as Namalwa, 
Akuba, P3HR-1, and Ly-46 could be 
induced to produce between 200 and 
600 units of interferon. Thus LDV/7 cells 
may be included in the group of cells 
which spontaneously produce high amounts 
of interferon. Even though Namalwa cells 
can be used to produce higher yields of in- 
terferon with Sendai virus or treatment with 



118 



SPONTANEOUS INTERFERON PRODUCTION 



Poly(I)poly(C) plus DEAE-dextran in the 
presence or absence of butyrate (5), the 
amount of interferon produced is com- 
parable to that spontaneously produced 
by LDV/7 or 8866-7 and LuKII cells 
(300-1000 units). It seems therefore that 
LDV/7 cells, capable of replicating high 
levels of transfer factor and immune RNA, 
could be valuable in obtaining high yields of 
interferon in large-scale production without 
the costs and the complexity of the use of 
an inducer. In addition LDV/7 cells may be 
used for cloning the interferon gene and for 
studies of gene expression. 

Our experiments in Fig. 1 showed that 
TPA, PHA-P, and LPS failed to increase 
the levels of interferon. 

The absence of increased interferon pro- 
duction after mitogen stimulation may be 
due to interferon inhibition of mitogenic re- 
sponses in LDV/7 cells. Szigeti et al, (15) 
showed that human leukocyte interferon at 
a concentration of 100 U/ml suppressed the 
mitogen (PHA)-induced lymphocyte mi- 
gration. They further showed that if the 
lymphocytes were pretreated with inter- 
feron prior to PHA activation, it signifi- 
cantly reduced the action of leukocyte mi- 
gration inhibitory factor. Thus the lack of 
increase in interiferon levels in the LDV/7 
supernatants in the presence of PHA or 
LPS may be due to the fact that LDV/7 cells 
are already proliferating cells and cannot be 
stimulated by PHA or LPS treatment. 
However, the action of PHA on stimulation 
of interferon production by T lymphocytes 
may be different, from the response of B 
cells like LDV/7 as suggested by Nathan et 
al. (14). 

The kinetics of spontaneous interferon 
production by cell cultures has not been 
fully elucidated. The data of Pickering ^/ aL 
(7) indicate that culture growth was found 
to correlate with the levels of interferon in 
cell lines spontaneously producing between 
300 and 1000 units. They based this finding 
on spontaneously produced interferon 
being elaborated continuously in seeded 
cells into fresh medium at early saturation 
of density and harvesting samples of 
medium after incubation of 1 or 2 days. The 
data in Fig. 1 also indicate that interferon 



production in LDV/7 cells is correlated with 
cell growth. Levels of interferon from Days 
1,3, and 7 were increasing and the samples 
from Days 10 and 14 showed that a plateau 
of spontaneously produced interferon was 
reached. Since the LDV/7 cells grow very 
rapidly (doubling time 24 hr) the cell sat- 
uration density occurs between 6 and 7 
days and immediately after the cells are in 
the lag phase. Thus these data further indi- 
cate that interferon was produced only 
during cell growth. 

The test of crude interferon against two 
oncogenic herpesvirus of primates indicate 
that it is active against HVS and EBV, sup- 
porting our previous findings where inter- 
feron obtained from both leukocyte and 
lymphoblastoid cells was more effective 
than fibroblast interferon (12). Even though 
P3HR-1 and Ag-876 expressing EBNA and 
EA/VCA, and Raji cells expressing only 
EBNA were not affected by the treatment 
with the LDV/7 interferon, these virus- 
infected cells could be destroyed by effec- 
tor cells activated by interferon (16, 17). 
Since the inhibition of transformation of 
cord blood cells by Ag-876 EBV was ob- 
served in low multiplicity of infection, 
perhaps more purified and higher doses of 
interferon may prevent transformation even 
with higher levels of viruses. On the other 
hand, the possibility of Ag-876 being less 
sensitive to interferon cannot be over- 
looked. 

The effectiveness of the interferon pro- 
duced by the LDV/7 cells in vitro against 
HVS and EBV provides promising support 
for in vivo trials against EBV- and HVS- 
induced tumors in subhuman primates. 

The possibility that this interferon has 
properties distinct from those of virus- 
induced lymphoblastoid interferon should 
be studied. Attempts will be made to in- 
crease the levels of interferon produced, 
either by examining clones of LDV/7 cells 
or by using other stimulating agents. 



I. Pizza G, Viza D, Boucheix C, Corrado F. In vitro 
production of a transfer factor specific for the cell 
carcinoma of the bladder. Brit J Cancer 
33:601-611, 1976. 



SPONTANEOUS INTERFERON PRODUCTION 



119 



2. Viza D, Boucheix C, Phillips J, Ccsarini JP, Phil- 
lips TM, Lewis MG, Pilch Y. Replication de 
rARN-immune par des cellules lymphoblastoides 
humaines. C R Acad Sci Paris 284:2565-2568, 
1977. 

3. Klein G, Dombos L, Gothoskar B. Sensitivity of 
Epstein -Barr virus (EBV) producer and non- 
producer human lymphoblastoid cell line to 
superinfection with EBvirus. Int J Cancer 
10:44-57, 1972. 

4. Pizza G, Viza D, Boucheix C, Corrado F. Effect 
of in vitro produced transfer factor on immune 
responses of cancer patients. Eur J Cancer 
13:917-924, 1977. 

5. Johnston MD. Enhanced production of interferon 
from human lymphoblastoid (Namalwa) cells 
pretreated with sodium butyrate. J Gen Virol 
50:191-194. 1980. 

6. Strander H, Mogensen KE, Kantell K. Production 
of human lymphoblastoid interferon. J Clin Mi- 
crobiol 1:116-117, 1975. 

7. Pickering LA. Kronenberg LH. Stewart W. 
Spontaneous production of human interferon. 
Proc Nat Acad Sci USA 77:5938-5942, 1980. 

8. Christofinis GJ, Steel CM, Finter NB. Interferon 
production by human lymphoblastoid cell lines of 
different origins. J Gen Virol 52:169-171. 1981. 

9. Viza D, Boucheix C, Cesarini JP, Ablashi DV, 
Armstrong G, Levine P, Pizza G. Characteriza- 
tion of a human lymphoblastoid cell line, LDV/7, 
used to replicate transfer factor and immune 
RNA. Biology of the Cell, 1982, in press. 



10. Green JA. Rapid assay of interferon. Texas Re- 
port on Biology and Medicine. 35:167- 172, 1977. 

11. Baron S, Dianzani F. General considerations of 
the interferon system. Tex Rep Biol Med 
35:1-10, 1977. 

12. Lvovsky E, Levine PH, Fuccillo D, Ablashi DV, 
Bengali ZH, Armstrong GR, Levy HB. Epstein- 
Barr virus and Herpesvirus saimiri: Sensitivity 
to interferon and interferon inducers. J Nat 
Cancer Inst 66:1013-1019, 1981. 

13. Pizzo PA, Magrath IT, Chattopadhyay SK, Biggar 
RJ and Gerber P. A new tumor-derived trans- 
forming strain of Epstein-Barr virus. Nature 
(London) 272:629-631, 1978. 

14. Nathan I, Groopman JE, Quan SG, Bercsh N, 
Golde DW. Immune {y) interferon produced by a 
human T lymphoblastoid cell line. Nature (Lon- 
don) 292:842-844, 1981. 

15. Szigety R, Masucci MG. Klein E, Klein G. Inter- 
feron suppresses antigen and mitogen induced 
leukocyte migration inhibition. Nature (London) 
288:594-5%, 1980. 

16. Adams A, Strander H. Cantell K. Sensitivity of 
the Epstein-Barr virus transformed human lym- 
phoid cell lines to interferon. J Gen Virol 38: 
207-217, 1975. 

17. Milfenhaus J, Damm M, Johannsen R. Sensitivity 
of various human lymphoblastoid cells to antiviral 
and anticellular activity of human leukocyte in- 
terferon. Arch Virol 54:271-277, 1977. 

Received April 15, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 120-125 (1982) 

Exercise and Estrus Cycle Influences on the Plasma Triglycerides of 

Female Rats (41487) 

WARREN K. PALMER* and JAMES R. DAVIS 

Department of Physical Education, University of Illinois at Chicago, Box 4348, Chicago, Illinois 60680 



Abstract. Since exercise and estrogens have a significant influence on plasma triglyceride 
(TG) concentration, this study was performed to determine the effects of exercise on the TG 
titers of female rats in the four stages of the estrus cycle. Normal female rats in the various 
phases of the estrus cycle, ovariectomized females, ovariectomized rats receiving estradiol, 
and normal male rats, all of comparable age, were run to exhaustion. At the time of exhaus- 
tion, the runner and a weight- matched control were anesthetized and exsanguinated. 
Ovariectomized animals receiving estrogen replacement ran 61% longer than the nude rats. 
However, this difference probably resulted from body weight differences, because when 
positive work was calculated, all group means were equivalent. Resting plasma TG levels 
were higher in normal male rats than in any other group. Ovariectomy had no effect on 
plasma TG levels but estrogen administration increased the concentration by 35%. Phase of 
the estrus cycle had no effect on resting TG levels. Exercise reduced plasma TG levels in all 
groups. The exercise-induced plasma TG response was not influenced by the phase of the 
estrus cycle. The concentration of TG at exhaustion was equivalent for all groups regardless 
of the preexercise TG level. These findings suggest that, during exercise, animals with high 
resting TG titers divert a greater portion of this fuel to oxidation than to tissue TG synthesis. 



Plasma triacylglycerols (TG) supply a por- 
tion of the fuel used by tissue for energy pro- 
duction (1). The muscle work of exercise in- 
creases energy utilization and fuel demand. 
It is well documented that exercise has a 
lowering effect on plasma TG. One bout of 
exercise will significantly lower the plasma 
TG level in humans (2) and rats (3). 

The ovarian steroid hormone estrogen 
has a striking, while paradoxical, influence 
on the level of TG in the blood. Removal of 
estrogenic hormones by ovariectomy or 
menopause is associated with an elevation 
in plasma TG titers (4). On the other hand, 
females receiving ovarian steroid hormones 
in the form of birth control pills have higher 
TG levels than females having normal 
menstrual cycle levels of estrogens (5). 
Moreover, rats receiving estrogen injec- 
tions also have circulating TG titers ele- 
vated above control levels (6). Since exer- 
cise and ovarian steroid hormones have a 
significant influence on plasma TG levels, it 
was the purpose of this study to determine 
the influence of an exhaustive bout of 



' To whom all correspondence should be addressed. 



treadmill running on the plasma TG levels 
in female rats having altered circulating 
levels of estrogens. 

Methods. Normal female, ovariec- 
tomized female, and male Sprague-Dawley 
rats (Harlan -Sprague-Dawley, Madison, 
Wise.) were used in this study. At the time 
of sacrifice all animals were 4 months of 
age. The mean body weight of each group is 
given in Table I. Since animals were age 
matched, it can be seen that there were 
mean differences in body weight with males 
being heavier than any other group. All 
animals were housed individually and pro- 
vided unrestricted access to Purina rat 
chow and water. The animal quarters were 
maintained at 23 ± T with a 12-hr light- 
dark cycle (0700- 1900 hr light). 

To determine the influence of removal of 
ovarian hormones, as well as the effect of 
estrogen replacement in ovariectomized 
rats on the parameters measured, surgical 
ovariectomy took place when the female 
animals were 3 months old. This was per- 
formed by removing the ovaries through a 
dorsal midline incision. Ovariectomy was 
characterized by a very small (67.4 ± 8.5 
mg) uterine weight at sacrifice when com- 



120 



0037-9727/82/090 1 20-06$0 1 .00/0 

? by the Society for Experimental Biology and Medicine. 



TRIGLYCERIDES IN EXERCISING FEMALES 



121 



TABLE I. Body Weights OF 
Females, 


Normal Females, Normal Males, Ovariectomized Sham-Injected (OSI) 
AND Ovariectomized Estrogen-Injected (OEI) Rats 


Normal 
females (64) 




OSI 

females (12) 


OEI 

females (12) 


Normal 
males (20) 


221 
±2.6 




248t 
±4.4 


216 
±3.1 


308* 

±2.7 



Note. All values are mean ± SEM; number of animals per group is given in parentheses. 

* Significantly heavier than all other groups {P < 0.05). 

t Significantly heavier than normal females and OEI females {P < 0.05). 



pared to normal control female uteri (140 ± 
8.2 mg). The mean body weight of these 
animals was also greater than normal. The 
day following surgery, 13 of the ovariec- 
tomized rats began receiving daily in- 
tramuscular hindlimb injections of 5 /Lig of 
estradiol-3 benzoate (Sigma Chemical Co.) 
in 0.2 ml of sesame oil (OEI group). The 
other 12 ovariectomized animals received 
daily 0.2 ml sham injections of sesame oil 
(OSI group). The mean uterine weight of 
the estrogen-injected animals was 314 ± 14 
mg at the time of sacrifice. 

All animals were indoctrinated to the 
treadmill with three 10-min running bouts. 
Only female animals were used that had 
normal estrus cycles, as determined by mi- 
croscopic evaluation of vaginal smears (7). 
This procedure was also utilized as an indi- 
rect index of the circulating ovarian hor- 
mone titers. Since the efTect of exercise on 
the plasma TG levels in males has been re- 
ported previously (3), we included a group 
of age-matched males in this study as a 
positive control. 

On the day of the exhaustive treadmill 
run, animals were weighed and vaginal 
smears were evaluated to determine their 
stage of the estrus cycle. Animals found to 
be in the same stage of the estrus cycle 
were paired by weight and assigned to 
either the runner or control group. Exer- 
cised animals were run on a Quinton 42-15 
rodent treadmill at 27.34 m/min up an 8% 
grade until the animals could no longer 
avoid the shocker. When the runner was 
removed from the treadmill, both runner 
and weight-matched control were ether 
anesthetized. The mean time of sacrifice of 
each group was similar to negate the circa- 
dian rhythm effect on the data. Rats were 



exsanguinated by cannulation of the ab- 
dominal aorta. Blood was collected in hepa- 
rinized tubes and centrifuged. Plasma was 
stored at -80^ until subsequent analysis for 
triglyceride concentration. Preliminary ex- 
periments indicated that freezing had no 
effect on the plasma TG values. The 
method of Fletcher (8) was used to deter- 
mine the amount of TG in each sample. The 
value obtained for each animal was the 
mean of duplicate analysis performed on 
two separate occasions. 

All values presented are group means ± 
SEM. Comparison of three or more means 
was performed using one-way analysis of 
variance. If an F ratio ofP< 0.05 was cal- 
culated, a Dunnett's post hoc test was per- 
formed (9). Two means were compared 
using Student's / test. 

Results. The mean run times of normal 
female rats at each phase of the estrus 
cycle, OSI females, OEI females, and nor- 
mal males are given in Table II. The run 
time of the ovariectomized rats receiving 
estrogen replacement was significantly 
greater than mean run times obtained for 
the male and normal females. When the 
phase of the estrus cycle was considered, 
the OEI run time was significantly greater 
than mean values obtained for runners in 
the estrus and proestrus phases of the es- 
trus cycle. 

The total amount of positive work per- 
formed for each animal was calculated from 
body weight, speed, and run time (up an 8% 
incline). The mean value for each group is 
found in Table III. It can be seen that no 
significant differences existed between 
group means. 

Mean plasma TG concentrations for each 
control and exercise group are given in 



122 



TRIGLYCERIDES IN EXERCISING FEMALES 



TABLE II. Run Times to Exhaustion for 
Normal Female Rats at Each Phase of the 
EsTRUs Cycle, Ovariectomized Sham-Injected 
(OSI) Females, Ovariectomized Estrogen- 
Injected (OEI) Females, and Normal Males 

Run times (min) 



Normal females 

Estnis (10) 

Metestrus(ll) 

Diestrus (10) 

Proestrus (6) 

Composite (37) 
(all normal female runners) 
Ovariectomized females 

OSI (6) 

OEI (6) 
Males (10) 



125.6 ±8.7 
144.3 ± 9.9 
135.5 ± 16.7 
98.8 ± 18.0 
129.5 ± 6.8 



141.5 ± 16.2 
177.8 ±21.3* 
110.3 ± 12.6 



Note. All values are means ± SEM; number of 
animals per group given in parentheses. 

* Significantly greater than the run times for aU the 
normal males, and all the normal females in the 
estrus and proestrus phases of estrus cycle. 



TABLE III. Positive Work" Performed for 
Normal Female Rats at Each Phase of the 
Estrus Cycle, Ovariectomized Sham-Injected 
(OSI) Females, Ovariectomized Estrogen- 
Injected (OEI) Females, and Normal Males 

Positive work 
performed (kg*m) 

Normal females 

Estrus (10) 60.6 ±4.4 

Metestrus(ll) 69.4 ± 4.8 

Diestrus (10) 68.5 ± 9.5 

Proestrus (6) 46.7 ± 7.6 

Composite (37) 63.1 ± 3.4 

Ovariectomized females 

OSI (6) 74.9 ±9.1 

OEI (6) 85.6 ± 2.0 

Males (10) 74.3 ± 9.0 

Note. All values are means ± SEM; the number of 
animals per group given in parentheses. 

" Positive work was calculated from the product of: 
body weight (kg) x speed (m/min) x run time (min) x 
grade (0.08). 



Table IV. When group control values were 
compared, results indicated that the ovari- 
ectomized female rats receiving estrogen 
had significantly higher concentrations of 
plasma TG than sham-injected ovariecto- 
mized rats, as well as the overall composite 
TG value for all the normal female rats. 
Male control plasma TG levels were signifi- 
cantly higher than all other control groups 



with the exception of the ovariectomized 
females receiving estrogen. Exercise pro- 
duced a significant reduction in the plasma 
TG levels of all groups except the females 
in the diestrus stage of the estrus cycle. 

The decrease in plasma TG as a result of 
treadmill running was determined by cal- 
culating the difference in plasma TG levels 
between each control and weight-matched 



TABLE IV. Plasma Triglyceride Concentrations for Control and Exhausted Male Rats. 
Normal Female Rats, Ovariectomized Female Rats (OSI). and Ovariectomized Female Rats 

Receiving Estrogen Injections (OEI) 

Plasma triglyceride (mg%) 



Controls 



Runners 



Normal females 

Estrus 

Metesirus 

Diestrus 

Proestrus 

Composite 
(All normal females) 
Ovariectomized females 

OSI 

OEI 
Males 



81 ±8.87(7)+§ 
71 ±7.8(8)+§ 
93 ± 7.7 (8)§ 
93 ± 9.7 (6)§ 
84±4.3(29)t§ 



92 ± 10.4 (6)t§ 
124 ± 16.7(6) 
129 ±6.6 (10) 



49 ± 2.5 (6)* 
47 ± 3.7 (9)* 
69 ± 12.7 (8) 
53 ±7.1 (?)♦ 
55 ± 4.2 (30)* 



51 ±6.6(6)* 

63 ± 5.7 (6)* 

64 ±7.0 (10)* 



Note. All values are means ± SEM; number of animals per group given in parentheses. 
* Significantly reduced plasma TG level compared to corresponding control value (P < 0.05). 
■*■ Significantly different from plasma TG level of OEI controls (P < 0.05). 
§ Significantly different from plasma TG level of male controls {P < 0.05). 



TRIGLYCERIDES IN EXERCISING FEMALES 



123 



Hi 



ft 

O 



70 



60 



50 



UJ 

O 

s 



ft 
< 

2 



UJ 
CO 

< 

UJ 

ft 
o 

UJ 

o 



30 _ 



20 _ 



10 _ 



i 



pa 



:::^:; 



OIESTRUS raOESTRUS 



COMTOSITE 
ALL FEMALES 



OVARIEC. MALES 
ESTRADIOL 



Fig. 1. The mean difference ± SEM of plasma triglyceride concentrations between runners and 
weight-matched controls is given for the following groups: males, ovariectomized females, ovariec- 
tomized females receiving estrogen, female rats in the various phases of the estrus cycle, and a compo- 
site value representing all normal female rats, (a) denotes a significant difference between the mean for 
the composite females and the group indicated. 



runner. These mean values for each group 
are presented in Fig. 1. The TG reduction in 
estrogen-injected ovariectomized animals 
was 61.6 ± 15.5 mg%, a reduction of 49%. 
This exercise-induced decrease is similar to 
that calculated for the males of 48% (61.9 ± 
8.8 mg% reduction). These decreases are 
significantly greater than the 20.4 mg% 
(35%) reduction computed for all the nor- 
mal females. 

Discussion. In the female rat, ovarian 
steroid hormones are responsible for the 
estrus cycle. The fluctuations that occur in 
estrogens and progesterone during this 4- to 
5-day period are thought to be responsible 
for various rhythms seen in female mam- 
malian physiology and biochemistry. Al- 
though the most pronounced influences are 
on uterine tissue, estrogenic rhythms have 
been demonstrated in body temperature 
(10), food intake (11), body weight (12), and 
voluntary running activity (13, 14). It has 
also been suggested that ovarian rhythms 
contribute to changes in blood hemoglobin 
content (15), plasma triglyceride titers (16), 
and tissue glycogen levels (17). 



Since most exercise-related research 
avoids ovarian cycling as an experimental 
variable by using males or by ignoring 
cycling altogether, we have attempted to 
determine the effect of the ovarian estrus 
cycle on the plasma TG response to 
exhaustive treadmill running. The mean 
exercise time of ovariectomized female rats 
receiving estrogen was significantly greater 
than the means of all other groups. These 
findings are consistent with reports (13, 14) 
indicating that elevated ovarian estrogen 
titers increased voluntary running activity. 
Unfortunately, the relationship between 
voluntary running activity and run time to 
exhaustion is not known. These two pa- 
rameters may not be related at all, since we 
found that stage of the estrus cycle had no 
effect on running endurance capacity. It 
would seem more reasonable to conclude 
that the increased run time in estrogen- 
treated rats was the result of group differ- 
ences in body weight. When total positive 
work was estimated, taking body weight 
into consideration, work performance was 
equivalent for all groups. 



124 



TRIGLYCERIDES IN EXERCISING FEMALES 



It is evident from these data that the sex 
difference in plasma TG reported for hu- 
mans (18), is also found in rats (19). The 
plasma TG level for male rats is approxi- 
mately 50% higher than that determined in 
all the normal cycling females. 

Estrogens have been shown to play a role 
in the metabolism of lipid (20, 21). How- 
ever, no effect of phase of the estrus cycle 
on plasma TG was evident in this study. 
These results support the work of Punno- 
nen (23) who found no effect of the 
menstrual cycle on plasma lipids. Unfortu- 
nately, Punnonen only studied three time 
points in the 28-day cycle. Two studies 
have reported that fluctuations in human 
TG (23) and low-density lipoproteins (24) 
exist during the menstrual cycle. However, 
differences in these parameters were not 
statistically significant. 

While ovariectomy had no effect on 
plasma TG titers, daily injection of 5 /Ltg of 
estradiol into ovariectomized rats signifi- 
cantly elevated the circulating lipid titers. 
Hamosh et al. (6) have shown that ovari- 
ectomy in rats had no effect on TG levels. 
This is in contrast to the works of others 
(25, 26), who report that prolonged removal 
of ovarian hormones, due to menopause or 
oophorectomy in humans, resulted in eleva- 
tions in the concentration of plasma TG. 

Ovarian hormone therapy has been used 
as a method of birth control. Numerous in- 
vestigators, using both humans (27, 28) and 
experimental animals (16, 29), have re- 
ported estrogen-induced increases in plas- 
ma TG levels. We have confirmed this find- 
ing, since TG levels were elevated 379f 
by estrogen administration to ovariec- 
tomized rats. 

The acute effect of exercise on the cir- 
culating level of TG in both man (30) and rat 
(3) is well documented. Recently, Terjung 
et al. (31) have shown that the uptake of 
plasma TG-derived fatty acids was in- 
creased in working skeletal muscle. This 
was supported by the results obtained in 
this study. Regardless of the experimental 
treatment, plasma TG levels were reduced 
as a resuh of one exhausting treadmill run. 
Phase of the estrus cycle did not quantita- 
tively alter this exercise-induced reduction. 



It is interesting to note, however, that even 
though the initial level of TG in the plasma 
differed between groups, the level reached 
at exhaustion for all groups was approxi- 
mately the same S6 mg%, a level similar to 
that reported by Reitman et al. (3) for 
animals that swam to exhaustion. Since aU 
groups did approximately the same amount 
of work (Table 111), these data (Fig. 1) 
suggest that significantly more energy may 
be derived from plasma TG in the working 
male and OEl female rats, than in the OSI 
and normal cycling females. 

We wish to thank Mrs. Mary Ann Fritsch for her 
assistance in the preparation of this manuscript. This 
work was partially funded by a grant from the UIC 
Research Board. 



1. Jones NL, Have! RJ. Metabolism of free fatty 
acids and chylomicron TG in the rat. Aoier J 
Physiol 213:824-828, 1%7. 

2. Oscai LB, Peterson JA, Bogard DL, Beck RJ, 
Rothermel BL. Normalization of serum triglyc> 
erides and lipoprotein electrophoretic patterns by 
exercise. Amer J Cardiol 30:775-780, 1972. 

3. Reitmann J, Baldwin KM, Holloszy JO. In> 
tramuscular triglyceride utilization by red, white, 
and intermediate skeletal muscle and heart during 
exhaustive exercise. Proc Soc Exp Biol Med 
142:628-631, 1973. 

4. Shibata H, Matsuzaki T, Hatano S. Relationship 
of relevant factors of atherosclerosis to meno- 
pause in Japanese women. Amer J Epidemiol 
109:420-424, 1979. 

5. Carlton M, Simons LA, Gibson JC. Vascular dis- 
ease risk factors in women with premature 
menopause. Horm Melab Res 12:208-211, 1980. 

6. Hamosh M, Hamosh P. The effect of estrogens on 
the lipoprotein lipase activity of rat adipose tissue. 
J Clin Invest 55:1132-1135, 1975. 

7. Turner CD, Bagnara JT. General Endocrinology, 
6th ed. Philadelphia, Saunders, pp 456-468. 1976. 

8. Fletcher MJ. A colorimetric method for estimating 
serum triglyceride. Clin Chim Acta 22:393-397. 
1%9. 

9. Steel ROD, Torrie JH. Principles and Procedures 
of Statistics. New York, McGraw-Hill, 1960. 

10. Abramson M, Torghele JR. Weight, temperature 
changes, and psychosomatic symtomology in re- 
lation to the menstrual cycle. Amer J Obstet 
Gynecol 81:223-232, 1%1. 

11. Attah MY. Besch EL. Estrus cycle variations of 
food and water intake in rats in the heart. J Appl 
Physiol 42:874-877. 1977. 



TRIGLYCERIDES IN EXERCISING FEMALES 



125 



1 2 . Landau T, Zucker I . Estrogenic regulation of body 
weight in the female rat. Horm Behav 7:29-39, 
1976. 

13. Asdell SA, Doornenbal H, Sperling GA. Sex 
steroid hormones and voluntary exercise in rats. J 
Reprod FertU 3:26-32. 1962. 

14. Furusawa K. Some evidence of hormonal control 
in activity of skeletal musculature in man. Rev 
Canad Biol 21:325-334. 1%2. 

15. Vellar OD. Changes in hemoglobin concentration 
and hematocrit during the menstrual cycle. Acta 
Obstet Gynecol Scand 53:242-246, 1974. 

16. Kim HJ, Kalkhoff RK. Sex steroid influence on 
triglyceride metabolism. J Clin Invest 56:888- 
8%, 1975. 

17. Palmer WK. Goldfarb AH. Ivy JL. Circadian and 
ovarian influences on tissue glycogen in female 
rats. Endocrinology 105:1254-1261, 1979. 

18. Hagerup L, Hansen PF, Skov F. Serum choles- 
terol, serum triglyceride and ABO blood groups 
in a population of 50 year old Danish men and 
women. Amer J Epidemiol 95:99-103, 1972. 

19. Bruckdorfer KR, Kang SS, Yudkin J. The hyper- 
lipidemic effect of sucrose in male and female 
rats. Proc Nutr Soc 31:1119-1120, 1972. 

20. Gerschberg H, Hulse M, Javier Z. Hypertri- 
glyceridemia during treatment with estrogen and 
oral contraceptives. Obstet Gynecol 31:186-189, 
1968. 

21. Meade TW, Haines AP, North WRS, Chakrabarti 
R, Horwarth PJ, Stiriing Y. Haemostatic, lipid, 
and blood-pressure profiles of women on oral 
contraceptives containing 50 fig or 30 /xg oestro- 
gen. Lancet 2:948-951, 1977. 

22. Punnonen R. Total serum cholesterol triglycerides 
and phospholipids during the normal menstrual 
cycle. Int J Gynaecol Obstet 15:2%-298, 1978. 

23. Kim H-J, Kalkhoff RK. Changes in lipoprotein 



composition during the menstrual cycle. Metabo- 
lism 28:663-668, 1979. 

24. Barclay M, Barclay RK, Skipski VP, Terebus- 
Kekish O, Mueller CH, Shah E, Elkins WL. 
Fluctuations in human serum lipoproteins during 
the normal menstrual cycle. Biochem J 96:205- 
209. 1%5. 

25. Hallberg L, Svanborg A. Cholesterol, phos- 
pholipids, and triglycerides in 50 year old women. 
Acta Med Scand 181:185-194, 1%7. 

26. Johansson BW, Kay L, Kullander S, Lenner HC, 
Svanborg L, Astedt B. On some late effects of 
bilateral oophorectomy, in the age range 15-30 
years. Acta Obstet Gynecol Scand 54:449-461, 
1%5. 

27. Corredor DG, Mendelsohn LV, Sabch G, Sunder 
JH, Danowski TS. Serum lipids during oral con- 
traceptive exposure. Clin Pharmacol Ther 
111:188-193, 1970. 

28. Hagopian M, Robinson RW. Effects of ethynyl- 
estradiol plus ethynylestradiol-3-methyl ether on 
serum triglyceride levels. Circulation 32 (Suppl. 
2): 14, 1%7. 

29. Schilling E. Gerhards E. Influence of hormonal 
contraceptives on carbohydrate and lipid metabo- 
lism in the rat. Biochem Pharmacol 22:1835- 
1844, 1973. 

30. Gyntelberg F, Brennan R, Holloszy JO, Schonfeld 
G, Rennie MJ, Weidman SW. Plasma triglyceride 
lowering by exercise despite increased food intake 
in patients with type IV hyperiipoproteinemia. 
Amer J Clin Nutr 30:716-720, 1977. 

31. Teijung RL, Budhoski L, Nazar K, Kobryn A, 
Kaciuba-Uscilko H. Chylomicron triglyceride 
metabolism in resting and exercising fed dogs. J 
AppI Physiol: Resp Environ Exercise Physiol 
52:815-820, 1982. 

Received April 19, 1982. P.S.E.B.M. 1982. Vol. 171. 



^fidendl 0/ the Societtf. 



Our members are requested to note the following list of Friends. We 
wish to express our thanics to each of them. Their contributions help 
materially in meeting the very sharp increase in costs of publication. 



FRIENDS 



Ciba Pharmaceutical Products, Inc. 

E. I. du Pont de Nemours & Co. 

Eli Lilly and Co. 

Hoffmann-LaRoche. Inc. 

Mead Johnson & Co. 

Merrill Dow Pharmaceuticals. Inc. 

Chas. Pfizer and Co.. Inc. 

Ortho Pharmaceutical Corporation 

A. H. Robins Co. 

Rogoff Foundation 

Sandoz Pharmaceuticals 



Schering Corporation 

Sharp and Dohme. Division of Merck & Co.. Inc. 

Smith Kline and French Laboratories 

Squibb Inst, for Medical Research 

Stuart Pharmaceuticals. Div. of ICI 
Americas. Inc. 

Upjohn Co. 

USV Pharmaceutical Corp. 

Wallace Laboratories. Division of Carter- 
Wallace. Inc. 

Wyeth Laboratories. Inc. 



NOTICE TO CONTRIBUTORS 

General Instructions 

Manuscripts should be written in clear, concise and grammatical English, and should conform to the general 
style of the Journal and the specific instructions listed below. Manuscripts which are not adequately prepared 
will be returned to the authors, since it is not feasible for the Editors to undertake extensive revision or rewriting 
of manuscripts submitted. Contributors, particulariy those unfamiliar with English usage, are encouraged to seek 
the help of colleagues in the preparation and review of manuscripts prior to submission. This practice will usually 
reduce the time required for review and will avoid delays in the publications of the manuscript. 

SUBMIT MANUSCRIPTS IN TRIPLICATE (ONE ORIGINAL AND TWO COPIES). A $10.00 CHECK OR 
MONEY ORDER (NON-REFUNDABLE) MUST ACCOMPANY THE MS TO COVER HANDLING COSTS 
FOR ALL MSS RECEIVED. Submitted manuscripts and accompanying materials become the property of the 
Society, and the Society is not responsible for loss or damage to such materials. 

All manuscripts should be submitted to Dr. M. R. Nocenti, 630 W. 168th Street, New York, N.Y. 10032. 

EDITORIAL OFFICE IS CLOSED DURING AUGUST. 

Authors submitting manuscripts containing data from experiments involving recombinant DNA molecules 
must provide a statement for each of the two reviewers which certifies that their experiments complied with the 
NIH guidelines on physical and biological containment procedures. 

1. Only original papers will be considered. Manuscripts are accepted for review with the understanding that 
the same work has not been and will not be published nor is presently submitted elsewhere, and that all persons 
listed as authors have given their approval for the submission of the paper; further, that any person cited as a 
source of personal communications has approved such citation. Written authorization may be required at the 
Editor*s discretion. Articles and any other material published in the Proceedings of the Society for Experimental 
Biology and Medicine represent the opinions of the author(s) and should not be construed to reflect the opinions 
of the Editor(s), the Society, or the Publisher. 

Authors submitting a manuscript do so on the understanding that if it is accepted for publication, copyright in 
the article, including the right to reproduce the article in all forms and media, shall be assigned exclusively to the 
Society. The Society will not refuse any reasonable request by the author for permission to reproduce any of his 
or her contributions to the journal. Send requests for permission to reproduce items published in Proceedings of 
the Society for Experimental Biology and Medicine to: Dr. Mero R. Nocenti, Editor, Society for Experimental 
Biology and Medicine, 630 W. 168th St., N.Y., N.Y. 10032. 

A manuscript rejected by the PSEBM should not be re-submitted. All manuscripts will be given a quality rating 
by the two reviewing editors; those manuscripts with low priority ratings will not be accepted even though they 
have been classed as generaUy acceptable. Split decisions will be decided on the basis of the two priority ratings 
and/or by a third review. 

2. a. Manuscripts should be as concise as possible, yet sufficiently detailed to permit critical appraisal. 

b. Manuscripts (including tables, legends, and footnotes) should be double or triple spaced. 

c. The first page of the manuscript shouki contain the complete title of the paper, category for the "Table of 
Contents" (select from list in item 20), names of authors (without degrees), affiliations (including Zip 
Codes), and a running title consisting of no more than 40 characters (including spaces). The second page of 
the manuscript should give the name, complete address, and telephone number of the author to whom ALL 
correspondence should be sent. Please include Zip Code. 

d. Units of weights, measures, etc., when used in coi\junction with numerals, should be abbreviated and 
unpunctuated, e.g., 6 R, 3 g, 5 ml, 8% (see No. 21 below). 

3. Manuscripts of nonexperimental researches, or those with inadequate controls, are not acceptable. 

4. Unnecessary subdivision of a research into several manuscripts is not acceptable. 

5. a. Manuscripts devoted to improvement of procedure or of apparatus may be accepted when a new 
principle is involved or when decidedly superior biological results are obtained. Evidence of such superi- 
ority should be given. 

b. Confirmatory or negative results will not be accepted unless they are of obvious biological significance. 

6. Manuscripts should not exceed 40 typed pages (including tables, charts, and references). 

7. Title should be limited to 15 words. Manuscripts should contain an Abstract, Introduction, Materials and 
Methods, Results and Discussion. Abstracts should not exceed 300 words. 

8. Conclusions should be based upon experimental data submitted. 

9. Figures. All figures should be cited consecutively by Arabic numerals in the text with figure legends typed 
on a separate sheet. These should contain sufficient experimental detail to permit the figure to be interpreted 
without reference to the text. Units should be clearly indicated in the figures themselves. Wherever possible, 
curves should be combined into a single figure in order to keep the number of illustrations to a minimum. 



PLEASE NOTE: All figures and illustrations are to be submitted in such form as to permit pbotographic 
reproduction without retouching or redrawing. This includes the lettering, which is reproduced as part of the 
photoengraving and is not set in type. Line drawings should be carefully drafted with black India ink on white 
drawing paper or blue drafting cloth, no larger than 8.5 x 1 1 .5 inches overall (21 x 27.5 cm). The lettering should 
be large enough to aUow a reduction of two-thirds. High quality glossy prints are acceptable. 

10. Tables. These should be numbered with Roman numerals and cited consecutively in the text. Each table 
should be titled and typed double-spaced on a separate sheet. Refer to current issues of the Proceedings for the 
acceptable style of tables. The title of each table should cleariy indicate the nature of the contents, and sufficient 
experimental detail should be included in footnotes to the entries to permit the reader to interpret the results. 
Units must be clearly indicated for each of the entries in the table. To save space, repetition of similar experi- 
ments and columns which can be calculated ftom other entries in the table should be avokled wherever possible. 

11. Footnotes. Footnotes in the text should be identified by superscripts consisting of Arabic numerals and 
should be typed on separate sheet: footnotes in the tables should be identified with superscript k>wer-case letters 
a, b, c, etc., and placed at the bottom of the table. 

12. a. References. Only essential citations should be submitted, and they should be arranged numericaUy at 
the end of the manuscript. References to the literature shoukl be cited in the text by Arabic numerals in 
parentheses, set on the text line. The references should be numbered consecutively in order of their 
citation. 

b. Abbreviations of journal titles should follow the style used in Chemical Abstracts (Vol. ACCESS, Key 
to the Source Literature of the Chemical Sciences, 1969 Edition). Please note the style of capitalization 
and punctuation for journal articles, books, and edited books in the following examples: 

1. Ludens JH, Bach RR, Williamson HE. Characteristics of the antinatriuretic action of 
growth hormone. Proc Soc Exp Biol Med 130:1156-1158, 1969. 

2. Abramson DI. Circulation in the Extremities. New York, Academic Press, pOOO, 1%7. 

3. Newell A. Simon HA. Programs as theories of higher mental processes. In: Stacy RW, Waxman ED, 
eds. Computers in Biochemical Research. New York, Academic Press, Vol 2:p(X)0, 1965. 

c. '^Personal communication,** ** unpublished,** '^submitted** and numerous abstracts should be excluded 
from the reference list. If the manuscript has been accepted for publication, include it in the reference list, 
giving journal, year, etc. If not accepted do not include it in the reference list. 

13. Trade or popular name or abbreviation of a chemical may be used only when preceded by the chemical or 
scientific name; thereafter, any of these names or abbreviations may be used. Trade names shouki begin with a 
capital letter. 

14. Structural formulas of chemicals should be used only when absolutely necessary. 

15. The Proceedings is copyrighted by the Society for Experimental Biology and Medicine. 

16. a. Authors are expected to discourage circulation of reprints for commercial purposes, 
b. Reprints are limited to 1300. 

17. Changes in galley proof should be absolutely minimal. Authors will be charged for excessive changes. 

18. If none of the authors is a member of the SEEM, the Ms must be sponsored by an SEBM member. The 
sponsor should write stating (a) how familiar he is with the author and his research, (b) whether the author is 
scientifically reliable, (c) that the manuscript conforms to Notice to Contributors, (d) that he has critically ex- 
amined the manuscript, (e) wherein the manuscript is a significant contribution to science, and (0 that he as- 
sumes financial responsibility involved. The sponsor should be in the same laboratory as the author. Where not 
of the same or recent past laboratory, he should specify why such letter comes fh>m a member at another laboratory. 

19. The authors are required to pay a part of the cost of publication in the form of a page charge of $20.00 per 



Notify the Executive Secretary at least one month before change of address. 

20. Biochemistry, Endocrinology, Growth/Development, Hematology, Microbiology/Immunology, Nutrition, 
Oncology, Pathology, Pharmacology, Physiology, Radiobiology, Tissue Culture. 

21. Abbreviations. Contributors are requested to use the following abbreviations: 



calorie 


cal 


centimeter 


cm 


counts per minute 


cpm 


cubic centimeter 


cm^ 


Curie 


Ci 


degree Celsius (Centigrade) 


-°C 


degree Fahrenheit 


op 



millimeter mm 

milliosmole mOsm 

minute min 

molal (concentration) m 

molar (concentration) M 

mole spell out 

molecular weight mol wt 



diameter 


diam 


nanogram 


ng 


gram 


8 


nanometer 


nm 


hour 


hr 


normal (concentration) 


N 


inch 


in 


osmole 


Osm 


inside diameter 


i.d. 


ounce 


oz 


intramuscular 


im 


outside diameter 


o.d. 


intraperitoneal 


ip 


parts per million 


ppm 


intravenous 


iv 


percent 


% 


kilocalorie 


kcal 


picogram 


Pg 


kilogram 


kg 


revolutions per minute 


rpm 


liter 


spell out 


second 


sec 


meter 


m 


specific activity 


sp act 


microliter 


Ml 


square centimeter 


cm^ 


micrometer 


fjitn 


square meter 


m* 


milligram 


mg 


subcutaneous 


sc 


milliliter 


ml 


volt 


V 






volume 


vol 



Rapid Communications 

22. Rapid Communications must be typed on special forms to be compatible with the camera-ready method 
of reproduction. The forms are to be obtained from the Editorial Office. The manuscript must not exceed four 
special form pages in length, including text, illustrative material, tables, and references. The style of the man- 
uscript, including the references, must conform to the style specified in the *' Notice to Contributors" for reg- 
ular manuscripts (see above). The manuscript must contain an Abstract, Introduction, Materials and Methods, 
Results, and Discussion. Type single-spaced with 12 pitch letters. The original and two copies of the manu- 
script are to be mailed to the Editor. All improperly prepared manuscripts will be returned to the authors. 

23. All Tables, Figures, or glossy prints must be on separate sheets, and their legends must be typed on sep- 
arate sheets also. Do not insert these items into the text typed on the special forms. Follow instructions in item 9, 
above. 

24. Each copy of the manuscript must be accompanied by a letter in which the authors justify the request 
for rapid publication. The objective of Rapid Communications is to provide rapid dissemination of timely, 
significant, and novel data of special or general interest. The novel aspects of data must be described clearly 
in the manuscript. 

25. Manuscripts submitted for the Rapid Communications section will be reviewed expeditiously by a single 
reviewer. It is anticipated that such manuscripts will be in print two months after their receipt. 



Advance Announcement 

of a forthcoming Serono Symposium 

on 

In vitro Fertilization 

October 25-26,1982 

Holiday Inn 

Carmel, California 

USA 



Scientific Committee 

D.T. Baird (GB), P.G. Crosignani (I), D.R. Mishell (USA), C. Pauerstein (USA) 

B. Rubin (I), R. Short (GB), M. Taymor (USA), D. Wood (AUS) 

Scientific Secretary 

P.G. Crosignani, IV Clinica Ostetrica e Ginecologica, Universita di Milano 

Via Macedonio Melloni, 52, 20129 Milan (Italy), Tel. (02) 74.04.21 

Main Topics 

• Sperm and Ovum Maturation 

• Fertilization 

• "In Vitro'' Models as Diagnostic Tests of Fertilizing Capacities 

• Embryo Cleavage, Ovum Transport and Implantation 

• Methods of Timing EGG Recovery 

• Clinical Experience of "In Vitro" Fertilization 

Useful Information 

Registration fee: US$300.00 or Lit. 300.000. 

The program does not include presentation of free communications. 

For further information please contact the Organizing Secretariat: 

Serono Symposia 

Via Ravenna, 8 

00161 Rome, Italy 

Tel. 06/423.328 Telex 680835 Serym 1 



A I • f jTmV ^ Subsidiary of Harcourt Brace Jovanovich, Publishers 

/\CQClGnilC ( >4r^) London New York Toronto Sydney San Francisco 

"^ 24-28 Oval Road, London NWl 7DX, England 

1 1 1 Fifth Avenue, New York, NY 10003, USA 

APL1268 




Pj.^^^ \y IX J 24-28 Oval Road, London NWl 7DX, England 



Society for Experimental Biology and Medicine 

630 WEST 168 STREET. NEW YORK. N. Y. 10032 
SUBMIT IN DUPLICATE 

Name of nominee Date 

(Pleue print last name fint) 

Mailing address Type of Nomination 

(Check one) 

Regular 

Associate 

Title now held Student 

Highest Degrees (give data and institution) 

Highest positions held in last 10 years in experimental biology and medicine, including post- 
doctoral fellowships (name of institution, title, dates) 



Honors, awards, etc. 



List membership in national scientific societies devoted to experimental biology and medicine 



The undersigned members of the Society agree that the above nominee is eligible for 

membership 

Signatures of 2 sponsoring members: 1 

2 



(over — for instructions) 



Eligibility 

Regular Membership: ''An applicant who has independently (not under direct super- 
vision of another scientist) published original meritorious investigations in experimental 
biology or experimental medicine . . . and who is actively engaged in experimental 
research may be elected to membership/' In general, applicants should be at least 2 years 
beyond their doctorate degree and beyond postdoctoral training. 

Associate Membership: ''Individuals who are engaged in research in experimental biol- 
ogy or medicine and/or teaching or administration in these areas. They must be sponsored 
by two regular members of the Society and must be residents of North America.** 

Student Membership: "Any student who is a candidate for a doctorate degree and is 
actively engaged in research in experimental biology or medicine, as attested to by two 
regular members of the Society (only one of whom needs to write a letter, but both need to 
sign the application), and who is a resident of North America.** 

Sponsorship. The application must be sponsored i.e., signed, by at least two members of 
the society. One detailed and factual letter from one of the member-sponsors describing 
the basis for eligibility must accompany this application. This letter should point out the 
contribution to science made by the applicant. Student applicants should have passed 
their qualifying examination and should be actively engaged in their thesis research proj- 
ect; date (month, year) on which doctorate degree is expected should be stated in the 
letter. 

Bibliography. Submit on separate sheets a complete list of publications, listing authors, 
title of publication in full, journal, year, volume and pages. Do not include articles not yet 
accepted for publication. Abstracts, reviews and books must be listed separately. 

Reprints. Submit one copy of each of three of your reprints. 

Address. Send this application in duplicate and supporting data to the Executive Secretary, 
Society for Experimental Biology and Medicine, 630 W. 168th Street, New York, N.Y. 
10032. Applicants will be notified of action taken as soon as the application has been pro- 
cessed (usually within a few months). 

Do not use this side of form. Use separate sheet if needed. 



In Two Volumes . . . 211227 

RADIOLOQICAL IMAQINQ 

THE THEORY Of IMAGE FORMATION, DETECTION, 
AND PROCESSING 

HARRISON H. BARRETT AND WILLIAM SWINDELL 
This two-volume work applies the analytical tools of modem 
image science — specifically, the theory of linear systems and 
stochastic processes — to radiological imaging systems that use 
high-energy photons. It covers the major imaging techniques: 
diagnostic radiology, nuclear medicine, and computed 
tomography. R^iohgicdl Imaging will be of interest to 
medical physicists, engineers, and physicists concerned with 
problems in imaging science and technology. It will also be 
of value to graduate students and researchers in the fields of 
imaging techniques in general and radiological imaging in 
particular. 

Contents of Volume 1 : THE CLINICAL SETTING. Conventional 
X Rays. Computed Tomography. Nuclear Imaging. Ultrasound. 
THEORY OF LINEAR SYSTEMS. Introduction. Impulse Response. 
The Frequency Domain. Linear Fillers. Sampling. Other 
Measures of Sharpness. THEORY OF RANDOM PROCESSES. 
Basic Concepts. Gaussian Random Variables. Poisson Random 
Processes. The Bernoulli Distribution. Filtering of Noisy 
Signals. APPLICATION OF LINEAR SYSTEMS THEORY TO 
RADIOGRAPHIC IMAGING. A General Model. Pinhole 
Imaging. Transmission Radiography. Scanners. Collimators for 
Scintillation Cameras. Further Complications. DETECTORS. 
Introduction. Film as a Detector. Film-Screen Combinations. 
The Gamma-Ray Camera. Non-Poisson Noise in Detectors. 
APPENDIX A: THE DIRAC DELTA FUNCTION. APPENDIX B: 
THE FOURIER TRANSFORM. APPENDIX C: INTERACTION OF 
PHOTONS WITH MATTER. APPENDIX D: RADIATION 
QUANTITIES. References. Author Index. Subject Index. 
Volume 1 : 1901, 384 pp., $55.00 ISiN: O-12-O7%01-5 

Contents of Volume 2: CLASSICAL TOMOGRAPHY. Principles 
of Classical Tomography. The PSF in Classical Motion 
Tomography. Variations on the Theme. Methods in Nuclear 
Medicine. COMPUTED TOMOGRAPHY. Introduction. 
Analytical Theory of Reconstructing from Projections. Practical 
Considerations. Emission Computed Tomography. MULTIPLEX 
TOMOGRAPHY. Introduction. The Annulus. The Fresnel Zone 
Plate. Pinhole Arrays. Time-Modulated Coded Apertures. 
THREE-DIMENSIONAL IMAGING. Introduction. Mathematical 
Preliminaries. Reconstruction by Back-Projection and 
Summation. Coordinate Transformations. Multiple-Aperture 
Systems. NOISE IN RADIOGRAPHIC IMAGES. Introduction. 
Summary of Basic Principles. Nuclear Medicine. Transmission 
Imaging. Computed Tomography. Coded Apertures. Noise and 
Image Quality. SCATTERED RADIATION. Introduction. 
Methods of Analysis for Scatter Problems. The Importance of 
Scattered Radiation. Control of Scattered Radiation. Uses of 
Scattered Radiation. References. Author Index. Subject Index. 
Volume 2: 1901, 352 pp., $55.00 ISiN: 0-12-079602-3 
Set price for Volumes 1 ainl 2 : $90.00 
($110.00, if purchased separately) 
Set prices are not valid in Australia or New Zealand. 

MICIIOFOCAL RADIOQIIAPHY 

EDITED BY R. V. ELY 

This authoritative volume traces the evolution of very fine 
focus x-ray generators and examines the potential advantages 
of higher resolution in the application of radiographic 
techniques. The chapters specifically draw attention to the 
importance of microconstituent behavior and highlight the 
inadequacy of some current radiographic procedures. The 
emphasis is on the wider applications of microfocal 
radiographic equipment and techniques. 
1901, 306 pp., $69.50 ISBN: 0-12-238140-8 

Send payment with order and save postage and handling. 
Prices »re in U.S. dolltrs »nd »re subject to change wit/iout notice. 

ACADEMIC PRESS, INC. 

A SubsidiMry of Harcourt Brace lovanovich. Publishers 
New York • London • Toronto • Sydney • San Francisco 
111 FIFTH AVENUE, NEW YOftK, N.Y. 10003 




OURCX)MPEnnON 
DNTINUALIY PURSUES THE MOST ADVANCE 
APPROACH TO BREEDING: ■ 

OURS. 




Some people still wonder what our 
competition will think of next. 
^ We don't. In fact, we're confident 
the innovations being perfected 
at Chailes River tociay will be 
imitated by our competitors 
for years to come. 
Take our breeding processes. 
We are the only commercial 
I breeder to offer Caesarean- 
Originated, Barrier-Sustained 
^. •^(COBS*) animals - so highly 
respected for ^ quality that some of our com- 
^titors have actually used them for starting their 
Am colonies. 

Charles River was the first to put all animal 
'oduction under direct veterinary supervision. 
We were also the first to pasteurize food and 
erilize bedding And the first to 
fer total lab support, critical to 
mtinuous health monitonng. 




Other Charles River firsts 
have taken place in virtually every 
aspect of our business. For research, we 
were the first to provide a virus-free inbred ; 
for production of monoclonal antibodies. 

In the area of customer ordermg and . 
inventories, we were the first to employ a comj 
erized Direct Order Entry System. As for produi 
innovation, we were the first commercial produ 
of free-ranging primates for research. Will our 
competitors try to copy this, too? We sue 

After all, monkey see. monkey do. 

Charles River Breeding Laboratories, ; 
251 Ballardvale Street, Wilmington, MA 01887, 
Telephone (617) 658^3333. 

Chailes River,^^-^' 

Breeding Labor atones, Inc. 



bred mgi 

amln^ 
/ a com J 

3r produi 
iai produ 
Villour 
ispec^ 

. Inc ™ 




l^cenii 



rssN m^iwm 



Volume 171, Number 2, November 1982 



PROCEEDINGS OF OEGH^ 

THE SOCIETY FOR --? 

Experimental 
Biology and 
Medicine 




ACADEMIC PRESS 

Ntw Yeri London 
Fftm Sun Dic^ Stin Fninci?u:t3 S^ Paulo Sydfiry 



Tokyo Toroolo 



Tfte Society was founded by S. J, Meltzer in 1903 



Even an idea 

whose time has cxxne 

needs help 

to make it ga 



The RolexAyvards for Enterprise. 



Rokx bebevei chat without proper binding 
many great ideas can never be fulfilled. So for the 
third time we are oflfering 50,000 Swiff fiana to 
eadiof five people with worthy projects in mind. 
To help their idieas become realities. In addition, 
each win receive a specially inscribed Gold Rokx 
Chronometer. 

Previous recipients have propoMd projects 
as varied as a program for ndio-cracking the 
endangered Himahyan snow leopard and a plan 
for doaimentation of European cave paintings. 
Yet each has shared the initiative and enterprise 
that have characterized Montres Rokx S.A. and 
the peopk who wear our watdi. 

Projects to be considered for the Awards 
should fall into one of three categories: 

Applied Sciences and Invention 
Exploration and Discovery 
The Environment 

A distinguished international committee 
has been sckc^ to judge die entries. 
SelectioB Committoc 
Andre J. Hriwij^fOwirifWi^— Chief Executive 
and Managing Director of Montres Rokx S.A., 
Switzerland 

Oriol Bchigas Guardiola —One of Spain s most 
important architects 

Walter Cunmngham—S former NASA astro- 
naut 

Professor Heinz Haber—Germin physicist and 
journalist 

Professor Mohamed /Gums— President of the 
lUCN and desert ccologist from Egypt 
Mrs. Patricia K{xrfc/in-5mytfcr— Former mem- 
ber of the British Equestrian Olympic Team, 



Tnsiee of die World Wildlife Fund iMcmaiional 
Jose M. Maycrga—A Spanish member of the 
Board of Trustees of die World Wildhfr IHuid 
International 

Professor Miduo NffM—Former Minister of Ed- 
ucation, Culture and Science for Japan 
Dr. Ruy Perez TtfiMyo— Head of Padiology at 
die Institute Nadonal de b Nutridon ""Sah^Mlor 
Zubiranr Mexico 
Mr. Paul'Emile Mfffor— French explorer and 



The appbcations win be judged on the basis 
of their general de m onstrations of the "spirit of 
enterprise" phis their quahties of newness, origi- 
nality, inventiveness, interest, feasibility, sigi^ 
ancc and likelihood of completion. 

Your entry must be completed in English 
on an official application form and reach the 
Secretariat before die 31st of March, 1963. The 
Awards win be presented in Geneva in April, 
1964. 

To obtain an official application form, to- 
gether with the detailed nilcs and conditions, 
write to: The Secretariat, The Rokx Awards for 
En^rise. RO. Box 178, 1211 Geneva 25, 
Switzerland. 

The Rolcx Awards for Enterprise. If you 
have an idea that s going pbccs, we an help it get 
there. 



ROLEX 




Examine the Rabbit Cage Units pictured 
above. 

Designed and manufactured by 
WARM ANN, they provide up to 600 
square inches of floor area and will accom- 
modate rabbits up to 12 pounds in weight. 

That's one reason this LC-321 Rabbit Unit 
ia a leader! 

Also, consider the heavy construction of 
the custom formed stainless steel stands. 
They are heli-arc welded and mounted on 
specially designed Wahmann LC-250 5" 
sealed bearing casters with brakes. 



Look beyond our Rabbit Units. Consider 
our other cages for your rats, mice, guinea 
pigs, dogs, cata, primates or poultry. 

They're all leaders.,. one of the reasons 
is.,. Because Wahmann is a privately, 
owned company with its sole operation 
being one of cage manufacturing and 
service, 

Wnte now for our new illustrated catalog. 

Wahmann Manufacturing Company 
10 West Aylesbury Rd* 
Tlmonium, Md. 2109^ 
Phone: (301) 252-2000 



m\ 



IM\ 



V% From the Comiiafty 

I ^Bsil^^%^^ that rs SPECIAU 

H^Sg^^^^. ^S fOf its CllStOfTWfS* 

%W Benefits 



Council 



President, Theodore Cooper *83 
The Ufyfohn Company 

President-Elect, Mariorie B. Zucker '83 
New York University 

• Past-President, Robert W. Berliner '83 
Yale University 

Treasurer, Harriet P. Dustan '84 
University of Alabama 

Executive Secretary, Mero R. Nocenti (ex officio) 
Columbia University 



Lloyd L. Anderson '83 
Iowa State Univ. 

R. AUERBACH '83 
Univ. of Wisconsin 

H. A. Bern '83 
Univ. of California 

Walter J. Bo '85 
Bowman-Gray Sch. of Med. 

M. G. Horning '83 
Baylor Coll. of Med. 



John H. Laragh '86 
Cornell Med. Ctr. 

Julian B. Marsh '85 
Med. Coll. Pennsylvania 

S. M. McCann '83 
Univ. of Texas 

John Resko '85 
Univ. of Oregon 

Wilbur H. Sawyer '86 
Columbia University 



Arthur A. Spector '86 
Univ. of Iowa 

J. Tepperman '83 

State Univ. of New York 

A. K. Weiss '83 
Univ. of Oklahoma 
Chairman, Membership Comm. 

M. B. Zucker '82 
New York Univ. 
Chairman, Publ. Comm. 



MEMBERSHIP APPLICATION 

Membership in the Society for Experimental Biology and Medicine is open to all individuals who 
have independently published original meritorious investigations in experimental biology or experi- 
mental medicine and who are actively engaged in experimental research. In general, applicants should 
be beyond a supervised postdoctoral experience in order to be able to demonstrate the ability to 
conduct independent investigations. 

Associate Membership is available for individuals who are engaged in research in experimental 
biology or medicine or in teaching or administration in these areas. Student Membership is available to 
any student who is a candidate for a doctorate degree and is actively engaged in research in experi- 
mental biology or medicine. Write to the Executive Secretary of the Society for details about eligibility 
requirements and application procedures for Associate and Student Memberships. 

Application forms may be obtained from the Office of the Executive Secretary, Society for Experi- 
mental Biology and Medicine, 630 W. 168th St., NY., N.Y. 10032. In addition, a tear-out applica- 
tion form is printed in each issue of the PSKBM. 



f*rocccdings of The StKicty for Experimental Biology and Medicine, Vol. 171, No. 2, November 1982. Published 
mi>nlhlv except August by Academic Press. Inc.. Ill Fifth Avenue. New York. NY. 10003. Second class postage 
paid at New York. NY and at additional mailing offices. 1982: Subscription per year S68.00 U.S.A.; $88.75 out- 
side USA. I98V Subscnption per year $73.(H) U.S.A. and Canada; $%.00 outside U.S.A. and Canada. All prices 
include postage and handling. Send notices of change of address to the Office of the Publisher at least 6-8 
weeks in advance. I*1ease include both the old and new addresses. Postmaster: Send all changes of address to 
PrtHeedinns of the StHieiv for Experimental Biology and Medicine. 11! Fifth Avenue, New York, N.Y. 10003. 
Copyright ^r^ 1982 by the SiKiely for F:xpcrimental Biology and Medicine. 



Board of Editors 



Editor 

M. R. NOCENTI 

630 W. 168th Street 

New York, N.Y. 10032 

212 795-9223 

Associate Editor 
Bernard F. Erlanger 

Columbia University 



UNIVERSITY eg 



oEcism 



Ahlquist 


Z. N. Gaut 


A. J. Marcus 


R. A. SCHEMMEL 


Allen 


A. L. Goldstein 


R. E. McCaa 


R. SCHMID 


.LisoN, Jr. 


E. C. GOTSCHLICH 


S. M. McCann 


N. J. Schmidt 


-PEN 


S. Greenberg 


J. C. McGiFF 


C. A. SCHNEYER 


Baehner 


M. R. C. Greenwood 


G. Medoff 


W. N. Scott 


Barraclough 


C. E. Grosvenor 


J. Meites 


J. E. Sealey 


Barron 


G. GUROFF 


T. C. Merigan, Jr. 


E. E. Selkurt 


Berg 


N. S. Halmi 


F. N. Miller 


L. Sensenbrenner 


)GGS 


M. A. Heymann 


C. R. Morgan 


J. H. Shaw 


3REL 


M. R. HiLLEMAN 


P. J. MULROW 


E. M. Shevach 


. Brand 


S. T. HOFFSTEIN 


L. H. MUSCHEL 


N. Shock 


tANSOME 


J. J. Holland 


G. D. NiSWENDER 


M. M. Sigel 


tESNICK 


J. A. HOLOWCZAK 


B. L. O'Dell 


G. W. Siskind 


Cain 


J. B. Hook 


S. Oparil 


J. L. Spivak 


Clark 


D. HORROBIN 


R. Ostwald 


M. B. Stemerman 


ark 


K. HUBNER 


P. Y. Paterson 


J. G. Stevens 


. Clifton 


E. D. Jacobson 


V. A. Pedrini 


B. J. Stover 


. Cohen 


H. D. Janowitz 


J. C. Penhos 


G. P. Studzinski 


. Cooper 


D. C. Johnson 


J. M. Phillips-Quagliata 


G. A. Tanner 


CORRADINO 


R. C. Johnson 


G. L. Plaa 


G. J. Thorbecke 


Cousins 


T. J. Kindt 


S. A. Plotkin 


M. L. Tyan 


Cremer 


F. G. Knox 


D. D. Porter 


J. L. Vaitukaitis 


DeLustro 


C. A. Krakower 


A. S. Rabson 


D. R. Van Campen 


Detwiler 


L. C. Krey 


J. A. Ramaley 


S. F. Vatner 


Salvo 


M. Kuschner 


R. J. Reiter 


C. M. Veneziale 


BE 


P. L. LaCelle 


J. A. Resko 


R. L. VicK 


CNOYAN 


C. A. Lang 


J. A. Rillema 


S. R. Wagle 


FiNKELSTEIN 


C. Lenfant 


R. B. Roberts 


M. E. Weksler 


Fisher 


R. Levere 


R. Ross 


J. M. Weller 


FOA 


J. V. Levy 


J. Rudick 


R. M. Welsh, Jr. 


FORKER 


G. LiTWACK 


T. M. Saba 


D. L. Wiegman 


. Frankel 


P. D. LOTLIKAR 


W. Sawyer 


D. B. Zilversmit 


Gala 


G. J. Macdonald 


B. B. Saxena 


B. G. Zimmerman 


Gallo 


I. Mandl 


A. J. Sbarra 


M. B. ZUCKER 




J. J. Marchalonis 


A. V. Schally 





Publication Committee 



M. ZucKER, '82, Chairman, M. Hilleman, '82; 
S. NiEwiARowsKi, '84; S. Seifter, '82; and J. Tepperman, '86, 
The President, President-Elect and Executive Secretary 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL 
BIOLOGY AND MEDICINE 

Volume 171, Number 2, November 1982 

Copyriglit © 1982 by the Society for Experimental Biology and Medicine 
All Rights Reserved 

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or 
mechanical, including photocopy, recording, or any information storage and retrieval system, without permis- 
sion in writing from the copyright owner. 

The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright 
owner*s consent that copies of the article may be made for personal or internal use. or for the personal or internal 
use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy 
fee through the Copyright Clearance Center, Inc. (21 Congress Street. Salem. Massachusetts 01970) for copying 
beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other 
kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating 
new collective works, or for resale. Copy fees for pre- 1982 anicles are the same as those shown for current 
articles. 

MADE IN THE UNITED STATES OF AMERICA 

Send requests for permission to reproduce items published in Proceedings of the Society for Experimental Biol- 
ogy and Medicine to: Dr. Mero R. Nocenti, Editor, Society for Experimental Biology and Medicine, 630 W. 168th 
St., New Yoric, NY 10032. 



INDEX TO ADVERTISERS 

Members and subscribers are requested to cooperate with our advertisers 
Charles River Cover 4 

Rolex Cover 2 

Wahmann Manufacturing Co fmi 

Camm Research Lab fmvi 



TABLE OF CONTENTS 

MINIREVIEW 

The Antigen Receptor of Thymus-Dcrivcd Lymphocytes: Pro- J. J. Marchalonis, J. C. Hunt 127 

gress in the Characterization of an Elusive Molecule 

ENDOCRINOLOGY 

Response of the Renal Vitamin D Endocrine System to A. D. Kenny. P. K. T. Pang 191 

Oxidized Parathyroid Hormone (1-34) 
Ovulation, Ovarian 17a-Hydroxylase Activity, and Serum M. S. Vomachka, D. C. Johnson 207 

Concentrations of Luteinizing Hormone, Estradiol, and 

Progesterone in Immature Rats with Diabetes Mellitus 

Induced by Streptozotocin 

GROWTH AND DEVELOPMENT 

Survival of Porcine Embryos after Asynchronus Transfer W. F. Pope, R. R. Maurer, F. Storm- 

SHAK 179 

HEMATOLOGY 

Macromegakaryocytosis After Hydroxyurea S. Ebbe, E. Phalen 151 

MICROBIOLOGY/IMMUNOLOGY 

Suppression of the Natural Killer Cell Activity of Murine W. I. Cox, N. J. Holbrook, R. J. Gras- 

Spleen Cell Cultures by Dexamethasone so, S. Specter, H. Friedman 146 

NUTRITION 

Effects of Garlic Products on Lipid Metabolism in Cholesterol- M. S. Chi 174 

Fed Rats 

ONCOLOGY 

Stimulation of Mammary Tumorigenesis and Suppression of H. Nagasawa, T. Mori 164 

Uterine Adenomyosis by Temporary Inhibition of Pituitary 
Prolactin Secretion during Youth in Mice 

PHARMACOLOGY 

Demonstration ofa^Adrenergic Receptors in Rat Pancreatic B. Cherksey, S. Mendelsohn, J. 

Islets Using Radioligand Binding Zadunaisky, N. Altszuler 1% 

PHYSIOLOGY 

Regional Uptake of pH]Norepinephrine by the Canine Left W. M. Chilian, R. B. Boatwright, 

Ventricle T. Shoji, D. M. Griggs, Jr 158 

Blood-Borne Vasoconstrictor Stimulates Release of ["C]- Y. Hazeyama, R. L. Moretti 168 

Arachidonate from Prelabeled Isolated Perfused Rabbit 

Gastrocnemius Muscle 

Evidence for an Intestinal Factor Stimulating Hepatic Cho- J. E. dos Santos, K.-J. Ho, C. L. 

lesterogenesis Krumdieck 184 

Myoelectric Activity of the Diverted Antroduodenum in J. Russell, P. Bass, A. Miyauchi 201 

the Dog 
Carbohydrate Metabolism during the Postprandial Intestinal R. H. Gallivan, Jr., C. C. Chou 214 

Hyperemia 

Golgi Complex Function in the Excretion of Renal Kallikrein E. Brandan, M. Rojas, N. Loyarte. 

F. Zambrano 221 

Erratum 

Volume 171. Number 4, September 1982: ''Annual Report of the Executive Secretary and Editor for 

the Year Ending December 31, 1981, ' pp. 523-524 232 



Highlights 
of Health 
Monitoring 
Reports 
on Camm 
Rats and 
Guinea Pig 



SEN. 



.0/10 



''should be considered 
truly pathogen-free/l 



^cofisf c(er«(( trufy pathogm free" 

^'no sendai virus in Camifi SPF mis" 

**na bacteriai respiratory pathosem'' 

**no murine poAogens.. . detected 
orisokted'* 

"no sdmomlla". . . **iio helminths** 

**na discltargefram nares^ canjunctiva 
ormus'' 

"entirefy suitable for my experimental 
purpMe" 



These excerpts from animal health mooitoring fse- 
ports on Camm rats and guinea pigs are tmly 
representative, although necessarily fragmented 
{ the independent monitoring laboratory is one of 
tiie most respected in the field. ) Please use the 
attached coupon and secure current, up-to-date 
feporls, You1l receive the full story of Camm ani- 
mals . . , bo(h the 'good news" and the "bad news" 
( if any. ) We want your sales and your r^pect. 

camm 

RESEARCH LAB ANIMALS 

414 BiKfc Otk RIdgt Rd,. Weyne, NJ 07470 Tel : (2(n) e9M)7(^ 



Ctifiifi^ own fleet of clim^e<ontroNed veftictes gives you express delivery rrom Boston to Wastiington 



CAWM RESEARCH LAB ANIMALS 

414 Black OaH Ridg^ Rd Wayne NJ 07470 



litest periodic health reports on: 
D Camm Guinea Pigs 
n SPF Rets 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 127-145 (1982) 



The Antigen Receptor of Thymus-Derived Lymphocytes: Progress in the 
Characterization of an Elusive Molecule (41488) 

JOHN J. MARCHALONIS^ and JEFFREY C. HUNT 

Department of Biochemistry , Medical University of South Carolina, 171 Ashley Avenue, 
Charleston, South Carolina 29425 



Abstract, Thymus-derived lymphocytes (T cells) show remarkable specificity in their 
capacity to recognize non-self antigens and this recognition must serve as the initial step in 
the differentiation of immunologically competent T cells into antigen-specific effector cells 
including helpers, suppressors, and cytotoxic lymphocytes. The problem of determining the 
molecular nature of the receptor for antigen on these cells is a challenging area of investiga- 
tion, and considerable insight into the serological and molecular properties of this receptor 
has recently been obtained using antibodies directed against immunoglobulin combining site 
regions as probes for the detection and isolation of the T-cell molecules. This review stresses 
results obtained within the past 3 years and (1) addresses the expression of immunoglobulin 
variable region determinants on T-cell receptors and factors, (2) presents a serological and 
molecular comparison of the structure of T-cell antigen-specific regulatory factors with 
those of receptors, and (3) presents a theoretical discussion of the genetics of antigen- 
specific T-cell factors and receptors. A pattern is emerging which indicates that T-cell 
receptors and some factors have a combining site which is related to immunoglobulin heavy 
chain variable regions. These molecules apparently do not bear determinants specified by 
the major histocompatibility complex (MHC), but express Ig-related variable regions and 
constant regions unique to T-cell products. The genes encoding these antigen-specific 
molecules (receptors, helper and suppressor factors) apparently are associated with the 
immunoglobulin heavy chain gene cluster. The intact VH-related T-cell molecules have a 
subunit mass of approximately 68,000 daltons and can form disulfide-bonded dimers. Studies 
using proteolytic enzymes, coupled with antigenic and functional analyses, indicate that the 
molecule is composed of domains resembling those of immunoglobulin heavy chains, al- 
though the T-cell molecule does not bear classical heavy chain isotypic determinants. The 
formation of active suppressor or helper factors often requires association of VH-related 
molecules with MHC-encoded proteins. 



Thymus-derived lymphocytes (T cells) receptor using antibodies directed against 
exhibit an exquisite specificity in their ca- combining site regions of immunoglobulins 
pacity to recognize antigen, and this recog- as probes for its detection, and for its isola- 
nition must serve as the initial step in the tion using immune affinity chromatography 
differentiation of immunologically compe- of T-cell receptor molecules. Antigen- 
tent T cells into antigen-specific effector specific T-cell immunoregulatory molecules 
cells such as helpers, suppressors, and such as helper and suppressor factors also 
cytotoxic lymphocytes. The problem of have been shown to express determinants 
determining the molecular nature of the serologically related to variable regions of 
surface receptor for antigen on T cells is a immunoglobulin heavy chains. A number of 
challenging area of investigation and has comprehensive reviews have been written 
been one of the major unresolved issues in regarding the problem of the T-cell receptor 
modem immunology. It has been possible for antigen (1-6). In this brief review, we 
to gain considerable insight into the sero- will focus predominantly upon data gener- 
logical and molecular properties of this ated in the past 3 years and consider the 

following major issues regarding the T-cell 
receptor for antigen: First, the evidence 

< To whom all correspondence should be addressed, supporting the existence of T-cell products 

127 

0037-9721/%2i VWiUl-VSl^^V .^^ 
Copyright ® \9K1 bv ^Ikve SocVevj \« 
All riihls reserved. 



128 



T-CELL ANTIGEN RECEPTORS 



related to immunoglobulin variable regions 
will be reviewed. Second, the relationship 
between T-cell receptors and T-cell re- 
leased effector factors will be analyzed. 
Third, recent data regarding the characteri- 
zation of isolated receptors and factors will 
be reviewed. Finally, based upon the mo- 
lecular characterization data presently avail- 
able, we will describe tentative models 
for the polypeptide structure of the T-cell 
receptor which is related to immunoglobu- 
lin heavy chain variable regions, and dis- 
cuss possible models for the arrangement of 
genes encoding variable and constant re- 
gions of T-cell receptors and their relation- 
ships to immunoglobulin heavy chain V and 
C genes. 

Existence of T-Cell Receptors and Fac- 
tors Serologically Related to Immuno- 
globulin Variable Regions. Although a 
number of studies performed in the early 
1970s indicated that some antisera direct- 
ed against immunoglobulin determinants 
reacted with T-cell products and could be 
used in the isolation of immunoglobulin- 
related T-cell surface markers (6-11) and 
factors (12, 13), the location of these deter- 
minants on the immunoglobulin-related 
molecule was not established. In retro- 
spect, it would appear that the original 
anti-immunoglobulin sera that were used 
to isolate immunoglobulin-related T-cell 
products must have done so because of 
a cross-reaction with variable region de- 
terminants because the ability of such sera 
to recognize determinants on T-cell mole- 
cules was not related to class-specific 
markers (6). Furthermore, subsequent 
studies established that such antisera were 
reactive with determinants lying in the Fd 
fragment of purified heavy chains (15), and 
also were associated with either interaction 
determinants formed by combination of V„ 
and Vl (16) or with restricted variable re- 
gion framework determinants (17). Two 
types of antisera have been very useful in 
establishing a relationship between sero- 
logical properties of the T-cell receptor 
and those of the combining site region of 
antibodies. The first type was the produc- 
tion of antibodies directed against idiotypic 
determinants on specific antibodies; the 



second type of reagent consisted of antisera 
directed against Vh framework determi- 
nants or to "non-idiotypic Vn-related de- 
terminants*' which are localized to some 
presently unspecified area of the variable 
region of heavy chains. Table I gives a par- 
tial listing of characterized idiotypic deter- 
minants which have been detected either on 
the surfaces of T cells or an antigen-specific 
T-cell factors. More than 13 defined 
idiotypic determinants have been found to 
be shared between antibodies and T cells or 
T-cell products exhibiting corresponding 
specificity. A range of idiotypic specificities 
have been detected which include (a) spec- 
ificity for defined low-molecular-weight 
haptens such as NP and the arsonate hap- 
ten, (b) specificities directed against natu- 
rally occurring proteins such as hen egg- 
white lysozyme, (c) specificities directed 
against synthetic polypeptides such as GAT 
and (T,G)A — L, (d) specificities for poly- 
saccharides, and (e) specificities directed 
against MHC alloantigens in mice and ro- 
dents. Moreover, idiotypes cross-reactive 
with those of human myeloma immuno- 
globulins have been detected on the sur- 
face of certain peripheral T cells in man 
and on isolated idiotype-bearing receptors 
(40, 41). The question whether or not the 
idiotype-bearing surface molecules or fac- 
tors detected were synthesized by the T 
cells has been answered affirmatively by a 
number of approaches (3, 4, 28, 30), and 
recent studies involving the production of 
T-cell hybridomas in vitro also strengthens 
the conclusion that T cells can synthesize 
and express idiotype related molecules (22, 
31, 42, 43). The range of defined speci- 
ficities and idiotypes of T cells suggests that, 
first, the recognition repertoire of antigen- 
specific T cells is diverse, and, second, that 
the antigen-specific products of T cells are 
serologically related in their combining site 
region to that of antibodies of correspond- 
ing specificities. It might be argued that the 
second property could arise by chance; that 
is, all proteins which bind a defined ligand 
with a certain affinity might be expected to 
have a similar geometry in the combining 
site and therefore have similar serological 
(or "idiotypic") properties. This argument 



T-CELL ANTIGEN RECEPTORS 



129 



TABLE I. Partial List of Idiotypes Shared between T Cells and Antibodies 
OF Corresponding Specificities 



"Common name** 




Reported occurrence on 


of idiotypic 












marker 


Specificity 


T cells 


T-cell factors 


NP 


(4-Hydroxy-3-nitrophenyl)acetyl 


Suppressor T-cell 


Antigen-specific 






hybrid: specific 


isolated receptor 






Thybridomas(104) 


(4, 18, 19) 




Timothy allergen 


Suppressors (20) 


Antigen-specific 
(20) helper factor 


HEL 


Hen eggwhite lysozyme 


Suppressors (21) 




(T.G)-A— L 


Tyramyl-glutamyl- 


Helper hybridomas 


Specific helper 




alanyl-lysine 


(22) 


factor (23) 


tyrtXMA) 


L-Tyrosine-p-azophenyl- 


Antigen-binding 






trimethylammonium 


T cells (24) 




ARSiAr) 


p-Azophenylarsonate 


Antigen-binding 


Specific suppressor 






T cells (25, 26) 


factors (27); 
biosynthetically 
labeled receptors 
(28, 29) 


R 5,936 


B6 anti-CBA antibodies 


AUoreactive 


Antigen-specific 




(MHC specific) 


T blasts (30) 


isolated receptor 
(30) 


GAT 


L-Glutamic acid ~-l- 




Specific suppressor 




alanine '^-L-tyrosine*® 




factor (31, 32) 




Human y globulin 




Specific suppressor 
factor (33) 


TEPC 15 


Phosphoryl choline 


Helpers (34, 35); 
delay ed-type 
hypersensitivity (36) 




A5A 


Streptococcal poly- 


Helpers (4, 18) 






saccharide A 


suppressors (4, 18) 




Nase 


Staphylococcal nuclease 


Helpers (37) 






Directed against combining 


AUoreactive 


Receptor (38, 39) 




sites of rat anti-MHC 


T cells (5) 






alloantibodies 






Idiotypes 


Unknown (40); 


Peripheral 


Isolated idiotype- 


of human 


anti-horse a,- 


T cells (40, 41) 


bearing receptors 


myeloma 


macroglobulin (41) 




(40, 41) 


immunoglobulins 









is essentially one of convergence in which 
molecules lacking common ancestral genes 
might have evolved similar structures be- 
cause of common function. The second ex- 
planation would be one of direct evolution- 
ary homology. Since hundreds of millions 
of years of evolutionary time were occupied 
in the generation of a variable gene genetic 
system to generate antibody diversity, it is 
reasonable to expect that antigen-specific 
lymphocyte products including antibodies 
and T-cell receptors would express com- 
bining sites encoded by the variable region 
genes. 

Possible examples indicating conver- 



gence have been reported, e.g., a shar- 
ing of idiotype between C-reactive protein 
and the phosphoryl choline-binding immu- 
noglobulin HOPC 8 (44), the unexpected 
cross-reaction between antibodies to the 
TEPC 15 Vk light chain and the Thy-1 al- 
loantigen (45), and the finding of a ubiqui- 
tous lymphocyte-associated protein which 
apparently shares the Ar cross-reactive 
idiotype as well as la antigenic determi- 
nants (46). However, the more reasonable 
conclusion at this time is that the T-cell re- 
ceptor molecules and factors express com- 
bining site determinants encoded by genes 
related to those of immunoglobulin heavy 



130 



T-CELL ANTIGEN RECEPTORS 



chain variable regions. This conclusion 
follows from the diversity of Vn-related 
idiotypes expressed by T cells. It appears 
possible that a sharing of 'idiotype*' might 
appear once or twice by chance; but it ap- 
pears inherently improbable that such an 
event would happen at least 14 times. A 
second point which indicates that the com- 
bining sites of antigen-specific T-cell prod- 
ucts most probably resemble immunoglob- 
ulin variable region-related idiotypes is that 
T cells recognize immunoglobulin idio- 
types, whether expressed on cells or on 
antibody molecules (47, 50). Current data 
show that helper T cells and their products 
are specific for antigen and express the 
idiotype of the corresponding antibody (19, 
23, 30), whereas suppressor T cells can be 
either specific for antigen and express 
idiotype (primary suppressors (27, 32)) or 
can react with idiotype, rather than with 
antigen (secondary suppressors (47-50)). 
Since idiotypes are usually formed as con- 
formational determinants which require 
interaction between Vj, and Vl(51-53), the 
fact that T cells and T-cell molecules (espe- 
cially products of secondary suppressor T 
cells) can react with conformational deter- 
minants further suggests that the T-cell re- 
ceptors, like antibody combining sites, rec- 
ognize three-dimensional shapes. This issue 
is worth considering because it has been re- 
ported that T-cell antigen receptors differ 
from antibodies in recognizing short, linear 
stretches of amino acids in denatured pro- 
teins, as shown in immune response gene 
effects expressed in macrophages (54). The 
property of recognizing short linear stretch- 
es of amino acids resembles properties of 
proteases (55), rather than those of an- 
tibody. 

The above data support a prima facie 
case for a sharing of idiotypic combining 
site determinants between antigen-specific 
T cells and their products, and antibodies of 
corresponding antigen specificities. In ad- 
dition, although T-cell receptors and factors 
have not been found to express any of the 
known immunoglobulin constant regions, 
they have been shown to share a number of 
other properties with those of defined im- 
munoglobulin variable regions. As of this 



time, there is persuasive evidence for a re- 
lationship between antigen-specific T-cell 
products and heavy chain variable regions, 
but the evidence for a relationship with light 
chain variable regions is much less certain. 
This particular problem is difficult to re- 
solve because many of the idiotypic deter- 
minants studied above require interaction 
with the proper light chain variable region 
to form the combining site. Other prop- 
erties shared between T-cell receptors and 
immunoglobulin heavy chain variable re- 
gions are as follows: association with al- 
lotype (30, 56), heterocliticity and fine 
structure for hapten binding (56), and 
nonidiotypic VH-related determinants (57- 
60). Cramer et al, (56) maintain that the 
Vh regions of T-cell antigen-specific re- 
ceptors share both framework and com- 
plementarity determining regions with Vh 
as expressed by B cells and antibodies. 

T-cell hybridomas have been constructed 
which produce suppressor factors specific 
for the protein keyhole limpit hemocyanin 
and express a membrane-associated an- 
tigen-specific receptor. The specific sup- 
pressor factors bear an antigenic determi- 
nant detected using rabbit antisera made 
against the Vh region of the murine my- 
eloma protein MOPC 315 (61). A point 
which will be considered below with re- 
spect to antigen-specific T-cell factors is the 
association of such functional molecules 
with products of the major histocompatibil- 
ity complex; a finding which has been ob- 
served in the case of keyhole limpit 
hemocyanin-specific suppressor factors 
produced by hybridomas, and in the case of 
the (T, G)-A — L specific helper factor and 
the GAT-specific suppressor factor de- 
scribed in Table I above. 

The present data indicate that the spec- 
trum of VfT^elated molecules expressed by 
T cells is most probably not identical to that 
of the entire Vh pool, but may represent a 
restricted subset of this pooL Evidence for 
this follows from the restricted expression 
of molecules related to the TEPC-15 
idiotype by T cells (35), and from the ex- 
pression of the NP idiotype by helper T 
cells and their factors. Apparently only Vh 
is required for the NP-idiotype of T cells. 



T-CELL ANTIGEN RECEPTORS 



131 



whereas the serum antibody requires both 
the presence of the proper V,^ and the 
proper Vh (56). In our hands, we find that 
some human T-cell tumor lines of amplifier 
phenotype express a VH-related determi- 
nant which comprises about 5% of the total 
human Vh pool and most probably is de- 
fined by amino acid sequence lying between 
residues 23 and the end of the heavy chain 
variable region (J. J. Marchalonis, J. C. 
Hunt, G. R. Vasta, unpublished observa- 
tions). 

Table II presents a comparison between 
properties of T-cell variable regions and 
variable regions of antibodies or B cells. As 
described above, a number of idiotypic de- 
terminants shared with antibody molecules 
have been described for murine T cells and 
T-cell products, and idiotypic receptors 
have been described on human T cells. The 
rabbit differs from man and mouse in ex- 
pressing allotypes in the variable region, 
and rabbit T-cell receptors have been de- 
scribed which bear the Vh allotype (62). It 
is worthwhile to note at this point that all 
investigators do not find Vn-related recep- 
tors on rabbit T cells. Jensenius et al, (63) 
do not find evidence for Vh (a allotype) 
markers occurring in the absence of light 
chains (b allotypes) by quantitative im- 
munoassay, although they find large (>10^ 
molecules/cell) numbers of Ig molecules in 
purified T-cell populations. They interpret 
their results to establish that Vh molecules 
found on T cells must represent B-cell 



contamination, and assert that all positive 
results obtained by other workers in any 
system must be due to either contamination 
of preparations with B cells or to use of 
poorly characterized antisera. Unfortu- 
nately, these workers present no data re- 
garding the properties of an alternative rec- 
ognition molecule. It is possible that the Vh 
determinant expressed on rabbit T cells and 
their products (62) is not a major a-allotype 
marker and that some antisera, thus, might 
not detect it. Furthermore, the rigid as- 
sumption that a-allotypes should be found 
on T cells only the complete absence of 
b-allotypes is unwarranted because the 
conformation of free Vh is different from 
that observed in the native state where it is 
noncovalently associated with V^ struc- 
tures. 

The murine T-cell receptor for the hapten 
NP resembles the NP-specific serum anti- 
body of the same strain in showing a 
heteroclitic response in which the hapten 
nitroiodophenyl binds better than the origi- 
nal NP immunogen. Heterocliticity, like 
idiot ype, is dependent upon the structure of 
the complementarity determining regions 
(hypervariable regions) within the antibody 
variable regions. In addition to idiotypic 
and heteroclitic properties, murine T cells 
and T-cell factors also express determi- 
nants which are associated with framework 
residues on murine heavy chain variable re- 
gions (56). Other determinants which can 
be localized to the heavy chain variable re- 



TABLE II. Comparison between Properties of T-Cell and Antibody (B Cell) Variable Regions 



Feature of T-cell 
"variable'' region 



Association with antibody 
variable regions 



(1) Idiotype (mouse, man) 

(2) Allotype (V„) 

(3) Heterocliticity (*Tme 
structure"); mouse 

(4) Framework, e.g., 
rabbit anti-mouse V|, 

(5) Nonidiotype; 
restricted determinant 

(6) Genetic linkage 
to Ch allotypes 



V„/V,^ interaction (52); V„ (56); 

complementarity determining regions (56) 
Framework sequences (78) variable 

D-J-Cm interaction (78); rabbit 
Complementarity determining regions 

(56), e.g.. anti-NP reacts better with NIP 

than with NP 
Framework sequences (79) 

Vh nonidiotype, location unknown: 

mouse (79). man (58-60) 
Genetic linkage to Ch allotypes 

(5, 56, 80) 



132 



T-CELL ANTIGEN BECEPTOBS 



gmi and which are not idiotypic have been 
described for T<ell products of moiMe and 
man ^57 59). Genen specifying idiotypic 
T'Cell variable regions have been found to 
he genetically linked to allotypes of immu- 
nogiobuiin constant regions fM-66i in a 
parallel fashion to that wdl established for 
Vh and Ch genes encoding immunoglobu- 
lins. 

This section, which is based upon the 
serological properties of the antigen- 
speciffc T'Cell receptor, indicates that T- 
cell antigen receptors possess a variable re- 
gion exhibiting remarkable similarity to 
antibody variable regions. However, it is 
still prnsiMe that T-cell variable regions are 
fHi4 identical to the V„ structures expressed 
by antibodies. T-cell variable regions may 
ftprtstnx a subset of the total V„ pool, or 
they mny represent molecules similar to the 
primitive variable regions in evolution, 
r»ther than being directly homologous to 
i^nUhiHiy Vh structures in higher species. 
Mr^f cover, although some T cells have been 
«ih/Fwn to prfKiuce messenger RNA for the 
cfFfistant region <ff fichaw, no evidence has 
yel heen published describing V/D/J/CH 
rfarr»n^ments in T cells (67-69). 

Hecepfors Versus Factors. Although 
nnllt',t'n s/frri/if %nlNhle effector factors 
all If h fncfllfile either helper or suppressor 
liint linn niltihf share the variable reffion 
f nnthhilnii site determinants with the 
intllt'rn sffct Ific T-i ell receptor and with 
fhr t niresffondinK antihtfdy, it does not 
ntt tssfifllv It flit fw that the constant rcf^ions 
ni f/lt't hn fnfrtlffn.s of the solithle mole- 
ctdrs Willi Id he identical to thifse (ff the 
I I I'll Miilm r iinli^en rcceittor. By analogy 
with IftifniifMij^lohulins, it would be cx- 
pcM fnl that the aiitijeen specific molecules 
of I ( rlls would lepiesent distinct classes 
(Isntvprs) depeuditi^ upon their function. 
(hM'ii and hei colleajeues (M 6()) have 
(ound fvidenre lot allotypes which appar- 
imiIIn aie associated with constant regions 
ol I cell deiixed molecules. These form a 
lamih id ihiee jzenes. iMie encoding a sup- 
pi rssiM lavtoi. one specilVing a helper T- 
vvW piodnci and the othei encoding the 
suilace lecepto! lound on immature lh\mic 
I lells ifvl i>(>\ (ieiuMic studies have 



shown tiiat these genes aire linked to senes 
spedfying anrine inmnoagkifaiiiin cooscaot 
regions and chat they map downstream 
from the lociis specifying Gz chains. A dif- 
ference between necepcoffs and fiKtors has 
been defined by functional analyses: 
namely, recotfmiion defined as binding of 
aniigen in soiution is nor dependent upon 
the MHC background of the T cells {70). 
whereas helper factors show a strong de- 
pendence upfin the la il-A \ background {23) 
and some, but not ail, antigen-specific sup- 
pressor factors require afiinctional associa- 
tion with products of I— J subregion {31, 
71, 72k 

It is useful to consider the amounts of T- 
cell antigen-specific membrane receptors 
and soluble fiftctors and to compare these 
quantities with other characterized mem- 
brane proteins found in lymphoid cells. 
Lymphocyte surface receptors usually 
comprise approximately 0. 1 to 1% of mem- 
brane protein. This figure has been ob- 
tained for histocompatibility antigens (73), 
the Thy-1 aOoantigen (74), and Virrelated 
receptors on primate T cells (75). The 
amount of a particular lymphocyte surface 
receptor computes to approximately 10,000 
per cell. Factors can function at extremely 
small protein concentrations. For example, 
the monoclonal T-cell suppressor factor spe- 
cific for GAT described by Knipen et al. 
(31) occurs at a calculated concentration of 
only 0.013 ng per mouse. This estimate rep- 
resents the amount of active idiotype- 
bearing factor produced by a small fraction 
of the heterogeneous murine T-cell pool. 
The yield recovered from the monoclonal 
hybridoma line allows an estimation of the 
amount produced per cell. Krupen et al. 
(31) isolated two micrograms of GAT- 
specific suppressor factor from 6 liters of 
culture fluid. Assuming that the cells 
reached a reasonable density of 10®/ml, and, 
using their molecular weight estimate of 
24,()(K) for the specific factor, each cell on 
the average is calculated to release ap- 
proximately 8(K)0 molecules. This amount is 
consistent with those noted above for iso- 
lation of membrane receptors and is also 
congruent with previous observations that 
antigen-specific T-cell receptors are "shed"* 



T-CELL ANTIGEN RECEPTORS 



133 



or released from the cell surface via a 
metabolic process (76) which differs from 
the secretory process carried out by acti- 
vated B cells or plasma cells. By contrast, a 
single plasma cell can secrete more than 
one million immunoglobulin molecules 
within an hour (77). These calculations il- 
lustrate the difficulty inherent in obtaining 
large amounts of T-cell receptor because 
monoclonal T-cell lines express and release 
only the same amounts of receptor which 
are associated with normal T cells. The 
hope of obtaining an immortalized T-cell 
line secreting large quantities (comparable 
to Ig secretion by plasma cells) has not yet 
been realized, although numerous antigen- 
specific monoclonal T-cell lines have been 
generated and studied (91). 

Despite the relatively low yields of T-cell 
receptors and factors which can be isolated 
even from monoclonal T-cell lines, a re- 
markable amount of serological and molec- 
ular information has been generated for 
T-cell receptors and factors in recent years 
as summarized in Table III. This table pre- 
sents data only on T-cell products which 
have been isolated by immune affinity 
chromatography and characterized to some 
degree by techniques such as polyacryl- 
amide gel electrophoresis. The first five 
listings might be classified as T-cell 
surface-associated receptors which bear 
either idiotypic markers or nonidiotypic Vh 
determinants. In these five separate cases, 
no MHC-associated products have been 
detected on the isolated receptor, and a 
common theme is evident in that molecules 
of approximately the size of heavy chain 
(50-70,000 d) are consistently isolated. The 
remaining items in the table consist of fac- 
tors which can be classified functionally as 
either antigen-specific helper or suppressor 
molecules which have been isolated either 
from sensitized normal T cells or from T- 
cell hybridomas. *'IgT" helper factors have 
been described in the mouse (12) and the rat 
(13). The nature of the immunoglobulin- 
related determinant on these molecules is 
not clear, but these helper factors do not 
carry MHC determinants. Other helper 
T-cell factors have been shown to carry 
both I- A and Vn-related determinants (23), 



or unspecified immunoglobulin determi- 
nants such as those detected using chicken 
antibodies directed against the murine /x 
chain (81). Lonai et al, (82) have generated 
murine T-cell hybridomas producing helper 
factor directed against chicken gamma- 
globulin, and this factor expresses Vh 
framework determinants detected using 
rabbit antiserum against the Vh of MOPC 
315, as well as I-A-associated determinants. 
A number of suppressor factors have re- 
cently been described; some of these con- 
sist of polypeptide chains of approximate 
mass 68,000-70,000 d and lack MHC com- 
ponents (84-89). On the other hand, the 
68,000-d suppressor factor specific for 
keyhole limpit hemocyanin which is pro- 
duced by a specific T-cell hybridoma de- 
scribed by Taniguchi et al, (61) apparently 
consists of two subunits. One of these has 
an approximate Mr of 45,000 d and bears Vh 
determinants whereas the other has an ap- 
proximate mass of 25,000 and bears I- J 
markers. Evidence indicates that these 
subunits can be covalently linked by disul- 
fide bonds in the ''secreted" form, or non- 
covalently associated in factors extracted 
by cell lysis. The GAT-specific suppressor 
factor isolated by Krupen et al, (31) con- 
sists of a single chain of Af^ 24,000 which is 
reported to carry both idiotypic and I-J 
markers. Although there has been consid- 
erable interest in the production and spe- 
cificity of factors directed against the arso- 
nate hapten because of the existence in A/J 
mice of a cross-reactive idiotype, the status 
of factors in this system is not clear. Both 
normal sensitized T cells (25, 26) and 
monoclonal T-cell hybridomas bearing the 
cross-reactive idiotype have been described 
(29), but confusion has resulted in regard to 
the nature of factors produced. One group 
reports the isolation, by affinity chromatog- 
raphy on the arsonate hapten, of a single 
chain molecule of mass 92,000 d which does 
not express Vh or MHC determinants (90), 
and another group reports the isolation of 
the single chain of approximate mass 62,000 
daltons which expresses both idiotype and 
I-J determinants (defined using alloantisera; 
29). However, the latter molecule is found 
in all lymphocytes (46). In our investiga- 



134 



T-CELL ANTIGEN RECEPTORS 



(2 

O 

z 
< 

o 



U 

H 

O 
w 

H 
< 



I 



0Q 
< 




5 
"S 



21 

1 




3 



III. 






3 



1 









Ills 



O O 

a a 
eg £ 






2 



I 



z 



2 



J. 
u 



z 



22 

CO 



2L 




2 



2 



C^. c^ c^. 



Z( 




< 



o 

I 



o 

Z 



Z2 



a. 
2 



•-I 

i ^= 

« ^ w J 



< 



a. 
2 



i> TT = C 



I w 2' ;j rfi 
I I- O H r4 






■o H -^ 13 ^ 

N B •'^ a. = 






a u s 



^■§ 



«* V. !i 

5 < 



o 

li 



CO CO 
<< 



a V 3 



E g.c; 



T-CELL ANTIGEN RECEPTORS 



135 



tions of an idiotype-bearing molecule 
biosynthetically produced by stimulated 
murine peripheral T cells, we have isolated, 
under reducing conditions, a single chain of 
68,000 d which also reacts with chicken 
antibody directed against the Fab fragment 
of murine immunoglobulin (26, 28). At this 
point in time, it can only be said that the 
results in the arsonate system are inconclu- 
sive and further analysis of antigen binding 
molecules is required, particularly because 
of the report that a widely distributed 
molecule of unknown function apparently 
binds arsonate and expresses a cross- 
reactive idiotype (46). 

Evidence obtained from murine (71) and 
primate systems (58, 59) indicates that a 
subpopulation of normal peripheral T cells 
(approximately 30% of PTL) expresses 
VH-related determinants (58, 59), and that 
this subpopulation also tends to express I 
region markers (71, 92). The studies sum- 
marized above indicate that the capacity of 
a T cell or a released molecule to combine 
with antigen results from the presence of a 
VH-related component, not from the pres- 
ence of an MHC marker. Because we have 
at hand human and lower primate T cell 
lines which express Vn-related surface 
components as well as HLA-Dr (the human 
equivalent of la markers), we carried out 
studies designed to determine whether or 
not the I region and VH-bearing components 
existed as a functional unit on the cell sur- 
face. Investigations involving direct isola- 
tion of either Vn-bearing (58) or HLA-Dr- 
bearing components (92) indicated that 
anti-VH antibodies isolated components of 
approximately 70,000 daltons, whereas 
monoclonal anti-la isolated components of 
28,000 and 32,000 d. We did not find evi- 
dence for a covalent or a strong noncova- 
lent association between these two compo- 
nents in the form in which they are released 
into the culture fluid or as they are ex- 
pressed on membrane fragments. In codis- 
tribution analyses using double immunoflu- 
orescence, we found a lack of congruity 
between HLA-Dr and Vn products (D. De- 
Luca and J. J. Marchalonis, unpublished 
observations). Although we cannot exclude 
the transient association of components of 



MHC and Vh systems, we did not find any 
evidence indicating a strong linkage be- 
tween the two sets of surface components. 

At this point in time, it is reasonable to 
conclude that the antigen-specific recogni- 
tion moiety on the T-cell receptor and re- 
leased effector factors is a V^-related 
marker, A good deal of evidence now 
suggests that this component most probably 
has an intact subunit mass of 68-70,000 d. 
This component binds antigen, and bears 
idiotype. Evidence also indicates that it (by 
itself) can bind to the surface of mac- 
rophages (1), presumably by some sort of 
constant region structure which can bind to 
a macrophage receptor (which is analogous 
to an Fc receptor). The association of Vn- 
bearing products with MHC products in the 
generation of helper or suppressor factors 
appears to be involved in situations which 
require cell/cell interaction, and might also 
be expected to depend upon the type of 
suppressor factor under consideration. For 
example, helper and suppressor factors 
would be expected to have different con- 
stant regions adapted for their particular 
effector functions, which would imply the 
existence of different MHC associations 
and different types of cell/cell interactions. 
Furthermore, it might be expected that 
primary suppressor factors (which bear 
idiotype and are antigen specific) could dif- 
fer from secondary suppressors (which are 
anti-idiotypic) which differ in their com- 
bining site specificity also could differ in 
MHC restrictions. This situation is analo- 
gous to that of antibody heavy chains where 
structurally and functionally distinct heavy 
chains (e.g., y chain and € chain) can share 
the same Vh structure. 

Tentative Model for Receptor and Fac- 
tor Structures. A number of significant is- 
sues remain to be resolved regarding the 
nature of T-cell receptors and factors. 
Among the most prominent of these are as 
follows: The valence or number of com- 
bining sites of intact T-cell antigen receptor 
in the absence of denaturing solvents is un- 
resolved. Since the isolated T-cell factors 
will neutralize hapten-derivatized bac- 
teriophage, the number of combining sites 
on the molecule or molecular complex must 



136 



T-CELL ANTIGEN RECEPTORS 



be at least two (93), and the size of these 
isolated receptors as estimated by gel fil- 
tration is approximately 150,000, a value 
which would correspond to a dimer of 
heavy chains. The requirement for light 
chain variable regions with the T-cell re- 
ceptor is unclear. Many of the antisera 
directed against idiotypes or Vh region de- 
terminants which react with T-cell receptor 
structures have a strong dependence upon 
the association of Vh and Vi^ for detection 
(51-53). In some cases, polypeptide chains 
resembling light chains have been isolated 
and partially characterized (1, 6, 14), al- 
though these molecules have been shown to 
be serologically distinct from standard k or 
X chains (17, 94) and they usually express a 
nominal molecular weight slightly higher 
than that usually observed for light chains. 
Third, the detailed molecular properties of 
T'Cell variable regions and their similarity 
to immunoglobulin variable regions both at 
the polypeptide and nucleic acid levels re- 
mains to be established. This is a funda- 
mental problem, and we will consider it in 
detail below. Even though antigen-specific 
T-cell products express a variable region 
serologically related to Ig Vh, the exact de- 
gree of homology can be answered only by 
detailed sequence analysis of the polypep- 
tide and its gene. 

Despite these questions to be resolved, a 
consensus regarding the properties of the 
Vn-related T-cell receptor is emerging from 
characterization studies being performed in 
many laboratories which were cited in 
Table III above. In particular, there is gen- 
eral agreement that T-cell receptors (and 
factors) bear serologically detectable vari- 
able regions and constant regions. Constant 
regions apparently unique to T-cell prod- 
ucts have now been detected using alloan- 
tisera (64-66, 95, 96), xenoantisera (97), 
and hybridoma antibodies (98) produced 
against isolated T-cell products of man and 
rodent species. The T-cell constant regions 
are distinct from the immunoglobulin 
isotypic determinants, but evidence now 
exists for the presence of a family of related 
isotypes of T-cell products (64-66, 95). 
These will be considered in detail below. It 
has recently become possible to isolate (by 



immune affinity chromatography) sufficient 
quantities of T-cell products (approxi- 
mately 100 /ig) to allow initial molecular 
characterization studies. Table IV presents 
a comparison of the amino acid composi- 
tions of Vn-related T-cell products isolated 
from a monoclonal T-cell hybridoma pro- 
ducing GAT-specific suppressor factor (31), 
a long-term in vitro grown marmoset T-cell 
line of amplifier phenotype (75, 99), and an 
idiotype bearing murine T-cell product (30). 
The molecules show an overall similarity, 
particularly in acidic amino acids, basic 
amino acids, and in the hydrophobic amino 
acids isoleucine, leucine, tyrosine, and 
phenylalanine. They are very similar in the 
hydroxy lie amino acid threonine. 

Sufficient quantities of the Vn-bearing 
T-cell products {receptor and certain fac- 
tors) have been isolated to allow charac- 
terization studies using standard tech- 
niques of protein chemistry in order to 
compare the structures of the T-cell 
molecules with one another and with clas- 
sical immunoglobulin chains. A number of 
serologically and functionally characterized 
fragments have recently been generated 



TABLE IV. Comparison of Amino Acid 
Compositions of VitRelated T-Cell Products 





Residues/ 100 residues 




Amino acid 


70-N2(T)- 


GAT-TsF* 


Tcr^ 


Asx 


12.7 


9.0 


10.1 


Thr 


5.1 


5.1 


5.4 


Ser 


10.5 


7.9 


6.7 


Glx 


14.8 


18.3 


12.8 


Pro 


6.7 


N.D. 


5.0 


Gly 


N.D. 


12.8 


6.5 


Ala 


10.5 


7.1 


8.1 


Val 


6.2 


5.5 


6.7 


Met 


0.7 


1.0 


2.5 


He 


3.1 


2.5 


3.0 


Leu 


7.5 


7.0 


9.5 


Tyr 


2.6 


4.0 


3.0 


Phe 


3.4 


3.3 


4.2 


His 


3.2 


2.5 


3.2 


Lys 


8.5 


7.4 


7.4 


Arg 


4.7 


6.6 


3.6 



" Data of J. J. Marchalonis, J. C. Maxwell, and C. Schwabe 
(unpublished observations) for V„-related product of the in 
vitro marmoset amplifier T-cell 70^N2. 

* Data of Krupcn et ai. (31) for GAT-specific suppressor 
factor from murine T-cell hybridoma. 

^ Data of Rubin et ai. (30) for the idiotype-bearing allo- 
specific murine T-cell receptor. 



T-CELL ANTIGEN RECEPTORS 



137 



(30, 39, 75, 86, 89) by cleavage with specific 
proteolytic enzymes or with CNBr. Figure 
1 presents data from this laboratory show- 
ing the intact 68,000-d product of a human 
T-cell line YT4E (Fig. lA, lane 1) and the 
cleavage products generated by tryptic 
proteolysis of this T-cell product (Fig. IB, 
lane 1) and of the corresponding product of 
the amplifier T-cell line 70-N2 (Fig. IB, lane 
2). Major fragments in the range 20-25,000 
d are generated as is a major fragment of 
approximate mass 45-47,000 d. In addi- 
tion, a number of higher-molecular-weight 
fragments are observed. The fragments in 
the molecular weight range 20-25,000 d 
react with antisera directed against Vh de- 
terminants; the major fragment of approxi- 
mate mass 45,000 d is not precipitated by 




Fig. 1. Analysis by polyacrylamide gel electropho- 
resis under reducing conditions of intact (A) Vh- 
associated T-cell product and fragments produced by 
tryptic proteolysis of the molecule (B). (A) Lane 1: the 
68,000-d component produced by the human in vitro 
grown T-cell leukemia line YT4E. This component 
was isolated from formic acid-solubilized membrane 
preparations by immune affmity chromatography 
using goat antisera directed against the Fab monomer 
fragment of a human IgM myeloma protein. Lane 2: 
molecular weight standards having masses as indicated 
on the gel. (B) Lane 1: tryptic fragments of the 
68,000-d VH-related product of the human T-cell 
lymphoma line YT4E. Lane 2: tryptic fragments of the 
68,000-d Vif-related molecule produced by the mar- 
moset in vitro T leukemia line of amplifier phenotype 
70-N2. V|i-related products were isolated and digested 
as described in ref. (75). 



anti-Vn reagents. These Vw-related prod- 
ucts of two separate T-cell lines are 
serologically related, but are not identical 
to one another. In addition (Fig. IB), their 
tryptic fragments give similar but not iden- 
tical patterns. Our data and those recently 
generated by other laboratories (30, 39, 86, 
89) are summarized in Fig. 2 which gives a 
schematic diagram illustrating observed 
fragments of Vh bearing T-cell receptors. 
The r chain undergoes a fragmentation 
pattern which indicates the presence of 
domains of approximate mass 12,000 d (30, 
75). A VH-bearing fragment (of approxi- 
mately 24,000 d) comparable to an Fd frag- 
ment of heavy chain has been observed 
(75), as has a fragment of about 45-47,000 
d (75, 89) which expresses effector func- 
tions similar to that shown by the Fc frag- 
ment of heavy chains (89). Notably, the 
Fc-like fragment of suppressor factors ap- 
parently expresses nonspecific suppressor 
activities (89). In addition, by following 
cleavage using CNBr with immune affinity 
chromatography it has been possible to 
isolate an antigenic fragment of approxi- 
mate mass 12,000 d which most probably 
corresponds to the T-cell heavy (r) chain 
variable region (75). In the cases which 
have been studied, the isolated r chains 
have had blocked N-terminuses (1, 30, 75), 
a result which might indicate some relation- 
ship to either the VhI or VhII subgroups of 
heavy chains. Rubin reports that his murine 
idiotype bearing T-cell receptor lacks 
detectable carbohydrate (30), but this 
question has to be resolved for other T-cell 
receptors and factors. Using papain diges- 
tion, we (J. C. Hunt, J. J. Marchalonis, un- 
published observations) have recently iso- 
lated a fragment of approximate mass 7800 
d which is hydrophobic, as assessed by 
elution behavior from reverse phase col- 
umns (by high-performance liquid chro- 
matography) and amino acid composition 
analysis. This peptide reacts with antisera 
directed against Fab region and Vh region 
determinants. It is possible that this frag- 
ment represents the disulfide-bonded loop 
of the variable region, and studies are in 
progress to establish its identity. The 
structure illustrated here is based upon 



tn 



T-CELL ANTIGEN BECEFTOBS 



^/" 



•»«r- 



fio.l. Schematic diagram iflu^tracing the proposed domain structure of V|«-beaniiK T-cefl receptors 
and the cvrpsniTkikm ofotnerved antigenic and functional fragments produced by proteolysis or chemi- 
cal cleavage. Thi^ diagram pertains only to the V^^^earing. non-MHC chain. It is Gkeiy that I-region 
products can he a^vKiated as separate chains with molecules of this nature in the formation of active 
fact^s or a possible recognition/activation complex on the ceD surface. Documentation regarding 
the^ fragments is given in the text. 



%tudic% of receptors and the suppressor 
factors which are of approximate mass 
68 70/)00 d and lack strongly associated 
MHC products. It is possible that either I-A 
or I'J region products are associated with 
functional T-cell factors either as a second 
polypeptide associated through noncova- 
lent linkage or through disulfide bonds. It is 
frequently found that products of separate 
chromosomes can form noncovalent mul- 
timcrs (hemoglobins) or disulflde-bonded 
assemblies (immunoglobulins). However, 
the question of a single polypeptide (31) ex- 
pressing both MHC-associated regions 
(chromosome 6 in man) and immunoglob- 
ulin associated regions (V„; chromosome 
14 in man) is a most interesting and unique 
one and requires further detailed analysis. 
Vonnihlc RclationMhips between T-Cell 
Receptor and Immunoglobulin Genes. 
Allhtm^h the smflof^ncal data marshalled 
ahnvr on idintsffcs and other V„ markers in- 
dh ates that anti^'en-specijh' T-rell reeep- 
tors /Hisses s a ( nndyininj^ site ref^ion show- 
inn lemarkidde similarity to immumtj^loha' 
lin y<ni(thle regions i'lahles I and 11 above), 
a /fiiradox remains hecanse no dejlnitive 
ret^orts have vet a/tpeared which establish 
that I I ells possess i:enes e(msistinf^' of 
leintani^etl \'n. />. ./, and immunoglobulin 
i II fei^ions {fi7 6V). It is now generally ac- 



cepted that T ceOs do not express standard 
immunoglobulin heavy chain constant re- 
gion isotypes, but it might intuitively be ex- 
pected that the heavy chain expressed by T 
cells would show the same sort of Vh, D, J, 
C rearrangement that immunoglobulins do. 
This arrangement has not been demon- 
strated, and it may be that either the VK-like 
products of T cells are not encoded by im- 
munoglobulin Vh genes, or that a different 
arrangement or type of rearrangement has 
occurred in the generation of T-cell prod- 
ucts. One important result of the serological 
studies which should be stressed here is 
that there are indications that the repertoire 
of V„ structures expressed by T cells most 
probably represents a subset of the total V„ 
population (35, 36), J, J. Marchalonis, J. C. 
Hunt, G. R. Vasta, A. C. Wang, unpub- 
lished observations). An important series of 
recent experiments directly germane to this 
question revealed constant region allotypes 
on products of T suppressor cells (64, 96), T 
helper cells (65, 96), and on T-cell primitive 
receptors (66). Owen and her colleagues 
have described a series of alleles for T-cell 
markers which map downstream from the 
C„ locus of the immunoglobulin chain clus- 
ter. Furthermore, Tokuhisa and Taniguchi 
(96) have also recently described two dis- 
tinct allotypic determinants on the antigen- 



T-CELL ANTIGEN RECEPTORS 



139 



specific suppressor and enhancing T-cell 
factors that are encoded by genes linked 
to the immunoglobulin heavy chain locus. 
Figure 3 presents a hypothetical model 
that delineates alternatives which might 
account for the observed observations. 
It has been proposed that the constant 
region of the T-cell receptor heavy chain, 
the T chain, might be very similar to the 
primitive heavy chain in immunoglobulin 
evolution (1, 100) and therefore the C^ 
gene might be located between the Vh 
cluster and the group of D genes. This loca- 
tion (position I), with either a different set 
of D-like or J-like genes or a lack of them, 
would account for present observations 
that V, D, J, and C rearrangements have not 
been found in T cells. Another major site on 
the immunoglobulin heavy chain chromo- 
some for the location of T-cell receptor 
genes is shown in position II which follows 
from the studies of Owen and her col- 
laborators (64-66). The constant regions of 
the suppressor, amplifier, and thymocyte 
antigen receptor are located to the right 
of the standard immunoglobulin isotype 
genes. This arrangement has been ques- 
tioned because, if standard V, D, J, and C 
rearrangement occurred within the T cells, 
it would be predicted that the entire collec- 
tion of immunoglobulin constant region genes 



would be deleted in T-cell maturation. This 
is not the case; in fact, T cells have C/x 
genes (101). It is possible, however, that a 
subset of Vh genes was duplicated and 
translocated to a position between the stan- 
dard Ch genes and the Q cluster. Because 
of the enormous interest and the vigor of 
the attack on the genetic location of T-cell 
receptor genes, it is anticipated that the 
exact solution to this problem will be found 
in the near future. 

Conclusions. The problem of the molec- 
ular nature of the T-cell receptor for antigen 
is one of the major unresolved issues in 
contemporary immunology. A consensus is 
developing regarding the existence of rec- 
ognition structures related to immunoglob- 
ulin heavy chain variable regions on the 
surface of certain functional T cells and on 
many of the properties of isolated receptor 
(and factor) structures. A molecule of ap- 
proximate subunit mass 68,000 d has been 
isolated from certain primate and rodent T 
cells and has been subjected to controlled 
proteolysis using various proteolytic en- 
zymes and standard chemical cleavage 
methods. This molecule can be degraded 
into a major fragment of Mr 24,000, which 
bears the Vh determinant and binds anti- 
gen, and to a fragment of approximate mass 
45-47,000 d which lacks Vh determinants 



^^}^ 




Fig. 3. Hypothetical model depicting possible arrangements of T-cell receptor variable and constant 
region genes within the immunoglobulin heavy chain gene cluster. The constant region of the T-cell 
heavy chain is designated C,. The markers Tu,/, T,ud. and T.^d are the T-cell allotypic constant region 
markers defined by Owen and her colleagues (64-66). I (?) and II (?) indicate possible positions at 
which the C, genes can be located. 



140 



T-CELL ANTIGEN RECEPTORS 



and carries out nonspecific effector func- 
tions. In addition, larger V^-bearing frag- 
ments as well as subfragments of the Vh 
have been isolated and partially charac- 
terized. Although this molecule does not 
share constant region determinants with 
immunoglobulin heavy chains, the present 
data suggest that the molecule resembles 
heavy chains in being formed of domains of 
approximate mass 12,000 d. Genetic map- 
ping studies indicate that the Vh structure 
associated with T-cell receptors maps with 
immunoglobulin Vh, although the genes en- 
coding T-cell Vh structures most probably 
represent a subset of the total Vh pool 
available to immunoglobulin heavy chains. 
It is also possible that the T-cell variable re- 
gions and constant regions represent direct 
lineal descendants of primitive immuno- 
globulins in evolution and, therefore, would 
show a closer sequence homology to im- 
munoglobulins of lower vertebrates rather 
than to those of man and rodents. Further 
amino acid sequence data and nucleic acid 
sequence data are required to test this hy- 
pothesis. The T-cell receptor molecule ap- 
parently possesses a blocked N-terminus. 
Although amino acid compositions have 
now been obtained for three Vn-bearing 
T-cell molecules, amino acid sequence data 
are lacking. Major issues which remain to 
be resolved are the direct demonstration 
using nucleic acid probes that the T-cell Vh 
genes lie within the immunoglobulin V„ 
cluster and the location of the constant re- 
gion genes of the antigen-specific T-cell re- 
ceptor. The genes that specify the factors 
likewise remain to be precisely mapped, 
although it is likely that antigen-specific re- 
ceptors and factors share the same Vh pool. 
One of the major features which is often 
taken to distinguish "recognition of anti- 
gen'' by T cells from that of antibodies and 
B cells is that the T-cell recognition process 
is usually considered to show restrictions 
imposed by the major histocompatibility 
complex whereas antibodies show no such 
constraints. Based upon the MHC restric- 
tion in T-cell recognition, it has been pro- 
posed that T cells either express one sur- 
face receptor capable of recognizing both 
MHC products and nominal antigens (single 



receptor), or two distinct receptors (dual 
recognition hypothesis), one recognizing 
antigen and the other recognizing MHC 
products. Antigen-specific T-cell hybrid- 
omas expressing two specificities for nom- 
inal antigen and two MHC backgrounds 
have been constructed in the attempt to 
determine whether or not Vh and MHC 
products assort independently in T-cell 
function (102, 103). The present results 
are inconclusive, however, because one 
group finds conjunct association (102), but 
the other provides evidence of indepen- 
dent functional distribution suggesting dual 
recognition (103). The molecular data re- 
viewed here suggest that the Vn-bearing/ 
non-MHC-associated structure most prob- 
ably binds antigen on the cell surface and is 
the antigen recognition unit in T-cell fac- 
tors, whereas an association between this 
molecule and 1-A- or I-J-specified products 
might be required for the production of ac- 
tive factors involved in cell/cell interac- 
tions. Substantial progress has been made 
in the molecular characterization of the elu- 
sive T-cell antigen receptor, although a 
formidable task remains in determining the 
exact mechanisms of antigen-driven activa- 
tion of T cells and cell/cell collaboration. 

Original work presented here was supported by 
Grant AI 17493 from the National Institute of Allergy 
and Infectious Diseases. We thank Mrs. Joan Maxwell 
for expert technical assistance. We thank Drs. M. 
Cramer. R. E. Cone. P. Lonai, B. Rubin, H. Wigzell, 
M. Taniguchi, T. Tada, and T. Kishamoto for provid- 
ing us with preprints of unpublished manuscripts. 

1 . Cone RE. Molecular basis for T lymphocyte rec- 
ognition of antigen. Prog Allergy 29:182-221. 
1981. 

2. Marchalonis JJ. The T cell antigen receptor: The 
minimal hypothesis revisited. Immunol Today 
3:10-17, 1982. 

3. Tada T, Okumura D. The role of antigen-specific 
T cell factors in the immune response. Advan 
Immunol 28:1-87, 1979. 

4. Rajewsky K, Eichmann K. Antigen receptors of 
T helper cells. Contemp Topics Immunobiol 
7:69- 12, 1977. 

5. Binz H, Wigzell H. Antigen-binding, idiotypic 
T-lymphocyte receptors. Contemp Topics Im- 
munobiol 7:113- 177, 1975. 

6. Marchalonis JJ, Cone RE. Biochemical and 



T-CELL ANTIGEN RECEPTORS 



141 



biological characteristics of lymphocyte surface 
immunoglobulin. Transplant Rev 14:3-49, 1973. 

7. Moroz C, Lahat N. Surface immunoglobulin of 
mouse thymus cells and its in vitro biosynthesis. 
Cell Immunol 13:397-406, 1974. 

8. Boylston AW, Watson SR, Anderson RL. 
Mouse T ceO tumor immunoglobulin. 1. Antigenic 
properties and effects on T cell responses. Im- 
munology 31:827-835, 1976. 

9. Putnam D, Clagett J, Storb U. Immunoglobulin 
synthesis by T cells. Quantitative aspects. J Im- 
munol 124:902-912. 1980. 

10. Singh G, Bazin H, Ladoulis C. In vitro synthesis 
and release of immunoglobulins by thymocytes 
of inbred rats. Mol Immunol 16:755-766, 1979. 

11. Hammeriing U, Pickel HG, Mack C, Masters D. 
Immunochemical study of an immunoglobulin- 
like molecule of murine T lymphocytes. Im- 
munochemistry 13:533-538, 1976. 

12. Feldmann M, Cone RE, Marchalonis JJ. Cell in- 
teractions in the immune response in vitro. VI. 
Mediation by T cell surface monomeric IgM. Cell 
Immunol 9:1-11, 1973. 

13. Taniguchi M, Tada T. Regulation of homocy to- 
tropic antibody formation in the rat. X. IgT-like 
molecule for the induction of homocytotropic 
antibody response. J Immunol 113:1757-1769, 
1974. 

14. Moseley JM, Marchalonis JJ, Harris AW, Pye J. 
Molecular properties of T lymphoma immuno- 
globulin. I. Serological and general physical 
chemical properties. J Immunogene 4:233-248, 
1977. 

15. Marchalonis JJ, Warr GW, Santucci LA, Szen- 
berg A, von Fellenberg R, Burckhardt JJ. The 
immunoglobulin-like T cell receptor IV. Quan- 
titative cellular assay and partial characterization 
of a heavy chain cross reactive with the Fd frag- 
ment of serum mu chain. Mol Immunol 17:985- 
999, 1980. 

16. Warr GW, Marton G, Szenberg A, Marchalonis 
JJ. Reactions of chicken antibodies with immu- 
noglobulins of mouse serum and T cells. Im- 
munochemistry 15:615-622, 1978. 

17. Cone RE, Rosenstein RW. Isolation of T cell 
membrane proteins (IgT) with antisera to non- 
isotypic determinants of immunoglobulins: Evi- 
dence that IgT 'Might'' chains are not identical to 
B cell kappa chains. Mol Immunol 18:67-77, 
1981. 

18. Eichman K. Expression and function of idio- 
types on lymphocytes. Advan Immunol 26: 
195-254, 1978. 

19. Cramer M. T cell factors-T cell receptors. In: 
Steinberg CM, Lefkovitz 1, eds. The Immune 
System: Basel. Karger, Vol. I: pp31 1-315, 1981. 

20. Malley A, Brandt CJ, Deppe LB. Preparation 



and characterization of the antiidiotypic prop- 
erties of rabbit anti-timothy antigen B helper 
factor and antimouse timothy IgE antisera. Im- 
munology 45:217-225, 1982. 

21. Harvey MA, Adorini L, Miller A, Sercarz EE. 
Lysozyme-induced T-suppressor cells and an- 
tibodies have a predominant idiotype. Nature 
(London) 281:594-5%, 1979. 

22. Eshhar Z, Apte RN, Lowry 1, Ben-Neriah Y, 
Givol D, Mozes E. T cell hybridoma bearing 
heavy chain variable region determinants pro- 
ducing (T, G)-A — L-specific helper factor. Na- 
ture (London) 286:270-272. 1980. 

23. Mozes E, Haimovich J. Antigen specific T cell 
helper factor cross reacts idiotypically with an- 
tibodies of the same specificity. Nature (London) 
278:56-57, 1979. 

24. Prange CA. Fiedler J, Nitecki DE, Bellone CJ. 
Inhibition of T-antigen-binding cells by idiotypic 
antisera. J Exp Med 146:766-778, 1977. 

25. Lewis GK, Goodman JW. Purification of func- 
tional determinant-specific idiotype-bearing 
murine T cells. J Exp Med 148:915-924, 1978. 

26. Warr GW, DeLuca D, Marchalonis JJ. The 
immunoglobulin-like T cell receptor. III. Binding 
of the arsonate hapten by idiotype-bearing T 
cells and antibody is blocked by avian antibody 
to (Fab')2. Mol Immunol 16:735-738, 1979. 

27. Nisonoff A, Greene Ml. Regulation through 
idiotypic determinants of the immune response 
to the p-azophenylarsonate hapten in strain A 
mice. In: Fougereau M, Dausset J, eds. Progress 
in Immunology IV. New York, Academic Press, 
PP58-80, 1980. 

28. Marchalonis JJ, Warr GW, Hunt JC. Wang AC, 
Decker JM. DeLuca D. Murine and primate T 
cell-derived proteins related to immunoglobulin 
combining site-region determinants. In: Jane way 
C, Sercarz EE, Wigzell H, eds. Immunoglobulin 
Idiotypes. New York, Academic Press, pp501- 
507. 1981. 

29. Pacifico A, Capra JD. T cell hybrids with arso- 
nate specificity. I. Initial characterization of 
antigen specific T cell products that bear a 
cross-reactive idiotype and determinants en- 
coded by the murine major histocompatibility 
complex. J Exp Med 152:1289-1301. 1980. 

30. Rubin B. Suzan M. Kahn-Peries B. Bayor C. 
Schiff C. Bourgois A. Isolation and biochemical 
characterization of idiotype-bearing T-cell re- 
ceptors which express private allotypes and/or 
isotypes: A review. Bull Inst Pasteur 78:305- 
346, 1980. 

3 1 . Krupen K. Araneo BA. Brink L. Kapp J A. Stem 
S, Weider KS, Webb DR. Purification and 
characterization of a monoclonal T-cell suppres- 



142 



T-CELL ANTIGEN RECEPTORS 



sor factor specific for poly (LGlu" LAla*»LTyr»«). 
Proc Nat Acad Sci USA 79:1254-1258, 1982. 

32. Germain RN, Ju ST, Kipps TJ, Benacerraf B, 
Dorf ME. Shared idiotypic determinants on an- 
tibodies and T-cell derived suppressor factor 
specific for the random terpolymer L-Glutamic 
acid "-L-Alanine»<'-L-Tyrosine»<'. J Exp Med 
149:613-622, 1979. 

33. Chaouat G. Human gammaglobulin tolerance: 
Shared idiotypy of T-cell suppressor factor and 
antibodies. Ann Immunol (Inst Pasteur) 131:91- 
95, 1980. 

34. Julius MH, Cosenza H, Augustin A A. Evidence 
for the endogenous production of T cell recep- 
tors bearing idiotypic determinants. Eur J Im- 
munol 8:484-491, 1978. 

35. Benca R, Quintans J, Kearney JF, Flood PM, 
Schreiber H. Studies on phosphorylcholine- 
specific T cell isotopes and idiotype-specific im- 
munity. Mol Immunol 17:823-831, 1980. 

36. Sugimura K, Kishimoto T, Maeda K, Yamamura 
Y. Demonstration of T15 idiotype-positive ef- 
fector and suppressor T cells for phosphoryl- 
choline-specific delayed-type hypersensitivity 
response in CBA/N or (CBA/N X Balb/C)F, 
male mice. Eur J Immunol 11:455-461, 1981. 

37. Nadler PI, Miller GG, Sachs DH, Hodes RJ. The 
expression and functional involvement of 
nuclease-specific idiotype on nuclease-primed 
helper T cells. Eur J Immunol 12:113-120, 1982. 

38. Binz H, Wigzell H. Shared idiotypic determi- 
nants on B and T lymphocytes reactive against 
the same antigenic determinants. V. Biochemical 
and serological characteristics of naturally oc- 
curring, soluble antigen-binding T-lymphocyte- 
derived molecules. Scand J Immunol 5:559-571, 
1976. 

39. Binz H, Wigzell H. T cell receptors with allo- 
major histocompatibility complex specificity 
from rat and mouse. Similarity of size, plasmin 
susceptibility and localization of antigen-binding 
region. J Exp Med 154:1261-1278, 1981. 

40. Lea T, Forre 0, Michaelsen TE. Natvig JB. 
Shared idiotypes on human peripheral blood B 
and T lymphocytes. J Immunol 122:2413-2417, 
1979. 

41. Preud'Homme JL, Klein M, Laboume S, Selig- 
mann M. Idiotype-bearing and antigen-binding 
receptors produced by blood T lymphocytes in a 
case of human myeloma. Eur J Immunol 
7:840-846, 1977. 

42. Kishimoto T, Suemura M, Sugimura K. Okada 
M. Nakanishi K, Yamamura Y. Characteriza- 
tions of T cell-derived immunoregulatory 
molecules from murine or human T hybridomas. 
In: Feldmann M, Schrier MH, eds. Lympho- 
kines. Vol 5, Monoclonal T Cells and Their 



Products. New York, Academic Press, ppl29- 
160, 1982. 

43. Kapp JA, Araneo BA, Clevinger BL. Suppres- 
sion of antibody and T cell proliferative re- 
sponses to L-glutamic acid 60-Lralanine 30-L-ty- 
rosine 10 (GAT) by a specific monoclonal T cell 
factor. J Exp Med 152:235-240, 1980. 

44. Volanakis JE, Kearney JF. Cross-reactivity be- 
tween C-reactive protein and idiotypic determi- 
nants on a phosphocholine-binding murine 
myeloma protein. J Exp Med 153:1604-1614, 
1981. 

45. Pillemer E, Weissman IL. A monoclonal anti- 
body that detects a Vk-TEPC15 idiotypic deter- 
minant cross-reactive with a Thy-1 determinant. 
J Exp Med 153:1068-1079, 1981. 

46. Clark AF, Capra JD. Ubiquitous nonimmuno- 
globulin p-azobenzenearsonate-binding molecules 
from lymphoid cells. J Exp Med 155:611-616, 
1982. 

47. Sherr DH, Ju ST, Dorf ME. Hapten-spccific T 
cell responses to 4-hydroxy-3-nitrophcnyl acetyl. 
XII. Fine specificity of anti-idiotypic suppressor 
T cells (TSi). J Exp Med 154:1382-1389, 1981. 

48. Minami M, Okuda K, Furusawa S, Benacerraf B, 
Dorf ME. Analysis of T cell hybridomas. I. 
Characterization of H-2 and Igh-restricted 
monoclonal suppressor factors. J Exp Med 
154:1390-1402, 1981. 

49. Sy MS, Nisonoff A, (jermain RN, Benacerraf B, 
Greene MI. Antigen- and receptor-driven regula- 
tory mechanisms VIII. Suppression of idiotype- 
negative, p-azobenzenearsonate-specific T cells 
results from the interaction of an idiotype- 
specific second order T suppressor cell with a 
cross-reactive-idiotype-positive, p-azobenzene- 
arsonate-primed T cell target. J Exp Med 153: 
1415-1525, 1981. 

50. Bona C, Paul WE. Cellular basis of regulation of 
the expression of idiotype. I. T cell suppressor 
cells specific for MOPC 460 idiotype regulate the 
expression of cells secreting anti-TNP antibodies 
bearing 460 idiotype. J Exp Med 149:592-600. 
1979. 

51. Sher A, Lord E, Cohn M. Reconstitution from 
subunits of the hapten binding sites and idiotypic 
determinants of mouse anti-phosphorylcholine 
myeloma proteins. J Immunol 107:1226-1234, 
1971. 

52. SchifF C, Boyer C, Milili M, Fougereau M. The 
idiotypy of the MOPC 173 (IgG„) mouse 
myeloma protein: Characterization of syngeneic, 
allogeneic and xenogeneic anti-idiotypic an- 
tibodies. Contribution of the H and L chains to 
the idiotypic determinants. Eur J Immunol 
9:831-841, 1979. 

53. Marchalonis JJ, Warr GW, Smith P, Begg GS, 



T-CELL ANTIGEN RECEPTORS 



143 



Morgan FJ. Structural and antigenic studies of an 
idiotype-bearing murine antibody to the arsonate 
hapten. Biochemistry 18:560-565. 1979. 

54. Rosenthal AS. Determinant selection and mac- 
rophage function. Immunol Today 3:33-34. 
1982. 

55. Barrett, AJ. McDonald. JK. Mammalian pro- 
teases: A glossary and bibliography. Vol I, En- 
dopeptidases. New York. Academic Press. 1980. 

56. Cramer M, Reth M, Grutzmann R. T cell Vh ver- 
sus B cell Vh. In: Janeway C, Sercarz EE, Wig- 
zell H, eds. Immunoglobulin Idiotypes. New 
York, Academic Press. pp429-439. 1981. 

57. Ben Neriah Y, Givol D, Lonai P. Simon MM. 
Eichmann K. Allotype-linked genetic control of 
a polymorphic Vh framework determinant on 
mouse T-helper cell receptors. Nature (London) 
285:257-259, 1980. 

58. Marchalonis JJ, Warr GW. Rodwell JD. Karush 
F. Surface component of primate thymus- 
derived lymphocytes related to a heavy chain 
variable region. Proc Nat Acad Sci USA 
77:3625-3629. 1980. 

59. F^rre 0. Michaelsen T, Natvig J. Nonidiotypic 
Vh antigens can be detected on human lympho- 
cytes. Clin Immunol Immunopathol 22:436-441, 
1982. 

60. Michaelsen TE. Lea T. Sheep, rabbit and chick- 
en antisera against a human Vh fragment: Re- 
activity with immunoglobulins and lymphocytes. 
Immunology 45:751-759, 1982. 

61. Taniguchi M, Saito T, Takei I, Kanno M, To- 
kuhisa T, Tomioka H. Suppressor T cell hy- 
bridomas and their soluble products. In: 
Feldmann M, Schrier MH. Lymphokines, Vol. 
5, Monoclonal T Cells and Their Products. New 
York. Academic Press. pp77-117, 1982. 

62. Krawinkel U, Cramer M, Mage RG, Kelus AS, 
Riyewsky K. Isolated hapten-binding receptors of 
sensitized lymphocytes II. Receptors from nylon 
wool-enriched rabbit T lymphocytes lack 
serological determinants of immunoglobulin con- 
stant domains but carry a locus allotypic marker. 
J Exp Med 146:192, 1977. 

64. Jcnsenius JC, Crone M, Koch C. The still 
elusive T cell receptor. On the possibility of a 
common V-gene pool for B- and T-cell-antigen 
receptor molecules. Scand J Immunol 14:693- 
704, 1981. 

64. Owen FL, Riblet R, Taylor BA. The T suppres- 
sor cell alloantigen Tsu^ maps near immunoglob- 
ulin allotype genes and may be a heavy chain 
constant-region marker on a T cell receptor. J 
Exp Med 153:801-810, 1981. 

65. Owen FL. Spurll GM. Evidence for a T cell con- 
stant region gene family: Characterization of cell 
surface antigens by immunoprecipitation with al- 



loantisera and monoclonal antibodies. In: Jane- 
way C. Sercarz EE.Wigzell H, eds. Immuno- 
globulin Idiotypes. New York. Academic Press. 
PP419-428. 1981. 

66. Owen FL. Spurll GM, Panageas E. Tth/*, a new 
thymocyte alloantigen linked to Igh-1. Implica- 
tions for a switch mechanism for T cell antigen 
receptors. J Exp Med 155:52-60, 1982. 

67. Kronenberg M, Davis MM, Early PW, Hood LE, 
Watson JD. Helper and killer T cells do not ex- 
press B cell immunoglobulin joining and constant 
region gene segments. J Exp Med 152:1745- 
1761, 1980. 

68. Kurosawa Y, vonBoehmer H, Haas W, Sakamo 
H, Trauneker A, Tonegawa S. Identification of D 
segments of immunoglobulin heavy-chain genes 
and their rearrangement in T lymphocytes. Na- 
ture (London) 290:565-570. 1981. 

69. Williamson AR. Genes coding for T-lymphocyte 
receptors. Immunol Today 3:68-72, 1982. 

70. Marchalonis JJ. Molecular interactions and rec- 
ognition specificity of surface receptors. Con- 
temp Topics Immunobiol 9:255-288, 1980. 

71. Tada T, Hayakawa K. Okumura K. Taniguchi 
M. Coexistence of variable region of immuno- 
globulin heavy chain and I region gene products 
on antigen-specific suppressor T cells and sup- 
pressor T cell factor. A minimal model of func- 
tional antigen receptor of T cells. Mol Immunol 
17:867-875. 1980. 

72. Taussig MJ. Review. Antigen-specific T-cell 
factors. Immunology 41:759-787, 1980. 

73. Robb RJ, Strominger JL, Mann DL. Rapid 
purification of detergent-solubilized HLA anti- 
gen by affinity chromatography employing anti- 
)82-niicroglobulin serum. J Biol Chem 251: 
5427-5428, 1976. 

74. Zwemer RK, Barstad PA, Acton RT. Isolation 
and characterization of murine cell surface com- 
ponents. I. Purification of milligram quantities of 
Thy-1.2. J Exp Med 146:986-1000, 1977. 

75. Marchalonis JJ, Hunt JC, Maxwell J, Wang AC. 
Antigenic polypeptide fragments of a receptor 
related to the Fab fragment of human immuno- 
globulin from thymus-derived lymphocytes. Proc 
Nat Acad Sci USA 79:4733-4736, 1982. 

76. Cone RE, Marchalonis JJ, Rolley RT. Lympho- 
cyte membrane dynamics. Metabolic release of 
cell surface proteins. J Exp Med 134:1373- 1384, 
1971. 

77. Marchalonis JJ, Nossal GJV. Electrophoretic 
analysis of antibody produced by single cells. 
Proc Nat Acad Sci USA 61:860-867, 1%8. 

78. Mage RG. A new look at the biological and ge- 
netic significance of rabbit heavy chain allo- 
types. Ann Immunol (Inst Pasteur) 130C:105- 
114, 1979. 



144 



T-CELL ANTIGEN RECEPTORS 



79. Bcn-Neriah Y, Wuilmart C, Lonai P, Givol D. 
Preparation and characterization of anti- 
framework antibodies to the heavy chain variable 
region (Vn) of mouse antibodies. Eur J Immunol 
8:797-801, 1978. 

80. Rubin B, Hertel-Wulff B, Kimura A. Alloanti- 
gen-specific idiotype-bearing receptors on 
mouse T lymphocytes. I. Specificity charac- 
terization and genetic association with the heavy 
chain IgG allotype. J Exp Med 150:307-321, 
1979. 

81. Howie S, Feldmann M. In vitro studies on H-2- 
linked unresponsiveness to synthetic polypep- 
tides. III. Production of an antigen-specific T 
helper cell factor to (T,G)-A — L. Eur J Immunol 
7:417-421, 1977. 

82. Lonai P, Puri J. Hammerling U. H-2-restricted 
antigen binding by a hybridoma clone that pro- 
duces antigen-specific helper factor. Proc Nat 
Acad Sci USA 78:549-553, 1981. 

83. Lonai P, Arman E, Savelkoul HPC, Friedman V, 
Puri J, Hammerling U. Factors, receptors and 
their ligands: Studies with H-2 restricted helper 
hybridoma clones. In: Fathman G, Fitch F, eds. 
Isolation, Characterization and Utilization of T 
Lymphocyte Clones. New York, Academic 
Press, in press. 

84. Cone RE, Rosenstein RW, Murray JH, Iverson 
GM, Ptak W, Gershon RK. Characterization of 
T-cell surface proteins bound by heterologous 
antisera to antigen-specific T-cell products. Proc 
Nat Acad Sci USA 78:6411-6415, 1981. 

85. Rosenstein RW, Murray JH, Cone RE, Ptak W. 
Iverson GM, Gershon RK. Isolation and 
characterization of an antigen specific, T-celi 
factor associated with the suppression of DTH. 
Proc Nat Acad Sci USA, in press. 

86. Cone RE, Murray JH, Rosenstein RW. Ptak W, 
Iversen GM, Gershon RK. The use of heteroan- 
tisera to T cell antigen binding proteins as probes 
for T cell receptors. In: Janeway C, Sercarz EE, 
Wigzell H, eds. Immunoglobulin Idiotypes. New 
York, Academic Press, pp457-482, 1981. 

87. Fresno M, McVay-Boudreau L, Nabel G, Cantor 
H. Biologic properties of a purified antigen- 
specific suppressive glycopeptide. Transplant 
Proc 13:1124-1127, 1981. 

88. Fresno M, McVay-Boudreau L. Nabel G, Cantor 
RH. Antigen-specific T lymphocyte clones. II. 
Purification and biological characterization of an 
antigen-specific suppressive protein synthesized 
by cloned T cells. J Exp Med 153:1260-1274, 
1971. 

89. Fresno M. McVay-Boudreau L, Cantor H. 
Antigen-specific T lymphocyte clones. III. Pa- 
pain splits purified T suppressor molecules into 



two functional domains. J Exp Med 155:981- 
993, 1982. 

90. Goodman JW. Lewis GK, Primi D. Horabeck P, 
Ruddle NH. Antigen-specific molecules from 
murine T lymphocytes and T cell hybridomas. 
Mol Immunol 17:933-945, 1980. 

91. Tada T, Nonaka M. T cell hybridomas and their 
products: An overview. In: Fitch F, Fathman G, 
eds. Isolation, Characterization and Utilization 
of T Lymphocyte Qones. New York, Academic 
Press, in press. 

92. Neubauer RH, Marchalonis JJ, Strand BC, 
Rabin H. Surface markers of primate B and T 
lymphoid cell lines defined by antibodies to 
human lymphocyte antigens. J Immunogenet 9: 
209-221, 1982. 

93. Cramer M, Krawinkel U. Immunochemical 
properties of isolated hapten-specific T cell re- 
ceptor molecules. In: Pemis B, Vogel HI, eds. 
Regulatory T Lymphocytes. New York, Aca- 
demic Press, pp39-55. 1980. 

94. Moseley JM, Beatty EA, Marchalonis JJ. Mo- 
lecular properties of T-lymphoma immunoglobu- 
lin III. Peptide composition of the light chain. J 
Immunogenet 6:19-28, 1979. 

95. Kontiainen S, Feldmann M. Structural charac- 
teristics of antigen-specific suppressor factors: 
Definition of "constant" region and "variable*' 
region determinants. Thymus 1:59-79, 1979. 

%. Tokuhisa T, Taniguchi M. Two distinct allotypic 
determinants on the antigen-specific suppressor 
and enhancing T cell factors that are encoded by 
genes linked to the immunoglobulin heavy chain 
locus. J Exp Med 155:126-139, 1982. 

97. Marchalonis JJ, Wang AC. A marmoset T- 
lymphocyte protein related to defined human 
serum immunoglobulin and fragments. J Im- 
munogenet 8:165- 175, 1981. 

98. Feldmann M, James R. Culbert E, Todd I, 
Makidono R. Cecka JM. Koniainen S. Uses of 
antifactor antisera and monoclonals in the 
analysis of T cell products. In: Janeway C, Ser- 
carz EE, Wigzell H, eds. Immunoglobulin 
Idiotypes. New York, Academic Press, pp483- 
492, 1981. 

99. Marchalonis JJ, Strelkauskas AJ. A marmoset T 
lymphoma which functions as a human amplifier 
T cell. J Immunogen 8:477-483, 1981. 

100. Marchalonis JJ, Warr GW, Vasta GR. Ledford 
BE. Molecular recognition of **non-self" deter- 
minants: The existence of a superfamily of rec- 
ognition molecules related to primordial immu- 
noglobulins. In: Gorczynski RM, ed. Receptors 
in Biological Systems. New York. Marcel Dek- 
ker, in press. 

101. Kemp DJ, Harris AW, Cory S. Adams JM. Ex- 



T-CELL ANTIGEN RECEPTORS 



145 



102. 



103. 



pression of the immunoglobulin C^ gene in 
mouse T and B lymphoid and myeloid cell lines. 
Proc Nat Acad Sci USA 77:2876-2880. 1980. 
Kappler JW, Skidmore B, White J, Marrack P. 
Antigen-inducible, H-2-restricted, interleukin- 
2-producing T cell hybridomas. Lack of inde- 
pendent antigen and H-2 recognition. J Exp Med 
153:1198-1214, 1981. 

Lonai P. Bitton S, Savelkoul HFJ, Puri J, Ham- 
meiiing GJ. Two separate genes regulate self-la 
and carrier recognition in H-2 restricted helper 



104. 



factors secreted by hybridoma cells. J Exp Med, 
in press. 

Tada T, Suzuki G, Hiramatsu K. Some com- 
ments on the antigen-receptors expressed on 
functional T cell hybridomas. In: Feldmann M, 
Schrier MH, eds. Lymphokines, Vol 5. Mono- 
clonal T Cells and Their Products. New York, 
Academic Press, ppll9-128, 1982. 



Received June 21. 1982. P.S.E.B.M. 1982. Vol. 171. 



fBOC££OfMM Of THE MCICT* fOB CJETEMME^-f AL WMMJOGt A.VD 



m. 



Suppression of the tiaiural Killer Cell Activity off Murine S p l een Cell Cultures tiy 

Dexamethasone (41489) 

WILLIAM L COX.* NIKKI J. HOLBROOK.^ ROBERT J. GRASSO.* 
STEVEN SPECTER.* and HERMAN FRIEDMAN*" 

* Department of Uedicai Ukroifioiof} amd immutmoio^ . CoMr^ cf Uedxime. L'mhmin 4ifSamtk Florida. 

Tampa, Florida 33612. amd ^Departmem cfFkxsiok^. Dmttmomtk Ue^cmi Schooi. 

Hamner. S'rm Hampskire 03755 



Abstract. Invcrtigattiot» imo the mode of actkwi of ghicocofticoid* o« oMaral kJertyK* 
ceO activity have been Inidered by the lack of aa in litro iMMki syslcoi. We fC|ion that the 
N'K activity of s'fecn ceO cuhnres of «e%-cfal iabred strains of nice mas sinnwrssed by- 
treatment with dexamethasone. The in ritro sopprcssioo of NK activity mas tiae depeBdent. 
requirmg at least 5 hr incubation in dexamethasone to achieve ntaiimal le%cls of sappres- 
sion, and was dose dependent at phannaro l n gi r concentrations. Thus, based on the resohs 
of these studies, an in vitro tuodei system for studying ghicocorticoid effects on NK activity 
has been established. 



Natural killer (NK) cells have been 
studied extensively in mice, rats, and man 
(1-4). The ability of these lymphoid cells to 
lyse many tumor cell lines without prior 
immunization implicates NK cells as a first 
line of defense against neoplasia. Recent 
evidence demonstrating that thymocytes 
may serve as targets of mouse NK activity 
suggest an additional role of NK cells in 
hemopoietic regulation (5). NK cell activity 
is enhanced by interferon itself or by sub- 
stances capable of inducing interferon 
(6-8). In contrast, NK activity is reported 
to be suppressed in vivo by glucocorticoids. 
Human peripheral NK activity of normal 
volunteers and patients with systemic lupus 
erythematosus is depressed by glucocor- 
ticoid treatment (9, 10). Similarly, rat and 
mouse NK activity is impaired severely 12 
to 24 hr after hydrocortisone administration 
(11, 12). It is not known whether glucocor- 
ticoids act directly or indirectly on NK cells 
and the lack of an in vitro model system has 
hampered such investigations. Although 
Hochman and Cudkowicz attribute sup- 
pressed NK activity in mice after hy- 
drocortisone treatment to stimulation of 
suppressor cells, more recent evidence 



' I () wlioiM all correspondence should he addressed. 



demonstrated that the spleen ceUs of cor- 
tisone acetate-treated mice, having de- 
pressed NK activity, fafled to suppress the 
NK activity of normal mice (13. 14). To 
understand the mechanism undertying glu- 
cocorticoid action on NK activity, the es- 
tablishment of an in vitro model system is 
essential. We report here the in vitro sup- 
pression of NK cell activity of mouse 
spleen cell cultures by pharmacologic con- 
centrations of dexamethasone. 

Mateiials and Methods. Mice. Inbred 
A/J (H-2^), BALB/cJ (H-y*), CBA/J (H-2''), 
C3H/HeJ (H-2»^), C57B1/6J (H-2»»), and 
DBA/2J (H-2**) mice were purchased from 
Jackson Laboratories (Bar Harbor, Maine). 
All mice used in these studies were males 
between 5 and 12 weeks of age. 

Target cells. YAC-1 lymphoma cells, 
previously described (15), were maintained 
in RPMl 1640 (Flow Laboratories, Rock- 
ville, Md.) medium supplemented with 10% 
fetal bovine serum (FLOW), 2 mAf l- 
glutamine (Grand Island Biological Co., 
Grand Island, N.Y.), 100 U/ml penicillin, 
and 100 ptg/ml streptomycin (GIBCO). 

Steroid treatment. Dexamethasone 
(Merck, Sharp and Dohme, West Point, 
Pa.) was initially prepared as a 10~"*Af stock 
solution in 95% ethanol and stored at 4°. 
For use in spleen cell cultures, 0.5 ml of the 



146 
^7 97J7/Hyi(H)l46'<m0l.m/i) 

right 't , /w.' hv the StKtety fnr l:xperimenlid Biology and Medicine. 
^0 /e*ef%>ed 



DEXAMETHASONE-SUPPRESSED NK CULTURES 



147 



Stock solution was evaporated in a 60 x 
IS-mm plastic tissue culture dish (Falcon, 
Oxnard, Calif.). The dexamethasone was 
redissolved in 5.0 ml of complete RPMI 
1640 (RPMI supplemented with 10% FBS, 2 
mM L-glutamine, 15 mAf HEPES, 5 x 10"* 
M 2-Mercaptoethanol, and antibiotics) by 
incubation at 37*^ for 30 min. Control 
medium was prepared in the same manner 
except that dexamethasone was omitted 
from the ethanol. Thus, both control and 
steroid preparations were virtually ethanol 
free. The pooled spleen cells of three to five 
mice were incubated at 37*^, 5% CO2 in 
RPMI 1640 containing dexamethasone or 
control medium. The cells were then 
washed twice in ice-cold RPMI 1640, 
counted in trypan blue with a hemacytom- 
eter, and adjusted to 10^ viable cells/ml. 

Cytotoxicity assay. Different effector to 
target cell ratios were prepared by twofold 
serial dilutions in complete RPMI 1640. NK 
activity was measured by a modification of 
the *^Cr release assay (16). Briefly, 0. 1 ml of 
YAC-1 target cells radiolabeled by incuba- 
tion in Na2**Cr04 (New England Nuclear, 
Cambridge, Mass.) were added to the indi- 
vidual wells of 96-well microtiter plates 
(Nunclon, Roskilde, Denmark) containing 
0.1 ml of spleen cell suspension. The plates 
were centrifuged at 250g for 2 min and then 
incubated for 4 hr at 37"" in a humidified 
atmosphere of 95% air, 5% CO2. The assays 
were terminated by centrifugation at 5(% 
for 10 min. Released radioactivity in 0.1 ml 
of the supernatant fluid was measured by 
liquid scintillation spectrometry. The per- 
centage cytotoxicity was computed from 

% cytotoxicity = (cpmexp - cpmgr)/ 
(cpm„,ax - cpmgp) X 100. 

cpmexp* cpm^ax* and cpmgr represent, re- 
spectively, the counts per minute in super- 
natants from YAC-1 target cells incubated 
with effector cells, from target cells lysed 
with 5% sodium dodecyl sulfate, and from 
target cells incubated without effector cells 
to give a measure of spontaneous release. 
Typically, spontaneous isotope release 
from target cells was less than 4% of the 
maximum isotope incorporation. Except 



where indicated, all results are expressed as 
the mean percentage cytotoxicity of at least 
three replicate cultures of one of several 
experiments. Standard deviations of means 
of the experiments shown were less than 
2% unless depicted. Three effector to target 
cell ratios, 25:1, 50:1, and 100:1 were used 
in all experiments and gave consistent re- 
sults. 

Experimental Results. Kinetics of sup- 
pression ofNK activity by dexamethasone. 
Mouse NK activity was measured in 5-hr 
chromium-51 release assays after preincu- 
bation of whole spleen cell preparations in 
dexamethasone, a potent synthetic glu- 
cocorticoid. Cultures that received con- 
trol medium demonstrated 10 to 11% 
cytotoxicity. In contrast, NK activity in 
10"^ M dexamethasone-treated cultures de- 
creased from 1 1 to 9% cytotoxicity after 1 
hr of incubation in medium containing the 
glucocorticoid (Fig. 1). By 6 hr of exposure 




HOURS AFTER ADDITION 

Fig. 1 . Kinetics of suppression of mouse NK activ- 
ity after in vitro cultivation in dexamethasone. 
C57B1/6 spleen cells were incubated in conical centri- 
fuge tubes at 37", 5% COj. At hourly intervals, control 
medium (O) or dexamethasone (•) was added to the 
appropriate cultures in volumes sufficient to yield final 
concentrations of 10~^ Af dexamethasone. After a total 
of 6 hr incubation, the spleen cells were washed and 
assessed for NK activity against **Cr-labeled YAC-1 
cells. Three effector to target cell ratios were used and 
gave consistent results. Shown here are the results for 
the 100:1 ratio of a representative experiment. Stan- 
dard deviations of the means of each point were less 
than 1% cytotoxicity. 



148 



DEXAMETHASONE-SUPPRESSED NK CULTURES 



to dexamethasone, NK activity had de- 
clined to 3% cytotoxicity, about one-third 
of the activity in control cultures. In three 
such experiments, the NK activity in cul- 
tures treated for 6 hr with 10"^ M dexa- 
methasone was significantly decreased 
compared to the NK activity of control 
cultures (P < 0.001, Student's t test). 

Experiments in which inhibitory con- 
centrations of dexamethasone were added 
directly into the cytotoxicity assay wells or 
in which YAC-1 target cells were cuhured 
for 24 hr prior to assay in medium contain- 
ing 10~^ M dexamethasone failed to demon- 
strate decreased levels of NK mediated 
cytotoxicity (data not shown). Thus, the 
suppression of NK activity in dexameth- 
asone-treated cultures is not attributable to 
the increased resistance of target cells to 
lysis after contamination of the cytotoxicity 
assays by possible carryover of dexametha- 
sone from the preincubation cultures, but 
to a dexamethasone effect on the spleen cells. 

Dexamethasone dose response. In order 
to assess the dose response relationship of 
the dexamethasone-induced in vitro sup- 
pression of NK cell activity, unfractionated 
spleen cells were incubated for 5 hr in vari- 
ous concentrations of dexamethasone be- 
fore assessment of NK cytotoxicity. The 
results of a representative experiment 
shown in Table I clearly demonstrate that 
incubation of unfractionated spleen cells in 
concentrations as low as 10~** M dexa- 



TABLK I. DosE-DhPhNDhNT in Vitro Suppression 
()^ NK Activity by Dexamethasone 





Percentage cytotoxicity 


DKX (M)« 


25* 


50 100 





6.1 


9.4 12.7 


10 " 


5.9 


8.4 13.2 


10 '" 


4.4 


7.0 10.2 


10 " 


1.4 


3.4 4.6 


10 « 


1.4 


2.0 3.1 


10 ' 


0.8 


1.5 2.0 


10 « 


0.8 


1.4 2.5 



" C'57BL'6J spleen cells ( lOVml) were incubated 5 hr in com- 
plete RPMl 1640 containing the indicated concentrations of 
dexamethasone. washed, and assayed for NK activity against 
VAC I target cells. 

" Ifleclor cell to target cell ratio. 



methasone was sufficient to induce dra- 
matic decreases in tai^t cell killing. Sup- 
pression of NK cytotoxicity was dose 
dependent from 10" M to lO"* M dexa- 
methasone. Increasing the concentration of 
dexamethasone to 10~^ M or 10~^ M did not 
increase the degree of suppression of NK 
activity. Thus, in vitro suppression of NK 
activity by dexamethasone is dose depen- 
dent and occurs at pharmacologic concen- 
trations. Similar results also were obtained 
from spleen cell suspensions of adre- 
nalectomized mice (data not shown). 

Dexamethasone suppression of NK ac- 
tivity of different mouse strains. The NK 
activity of inbred mice of different genetic 
backgrounds was tested for sensitivity to in 
vitro dexamethasone treatment. The results 
of a representative experiment, Fig. 2, 
demonstrate that the NK activity of all but 
one strain tested was strikingly suppressed 
after 5 hr incubation in medium containing 
10"^ Af dexamethasone. The exception, the 
spleen cells of A/J mice, had very little NK 
activity. Consequently, the suppressive 
effect of dexamethasone may not have been 
as apparent as that seen in the other strains 
that expressed much higher levels of NK 
activity. Nonetheless, it is clear that iVi vitro 
dexamethasone treatment suppresses 



> 

O 

s 

I- 
> 



n 



u I— Clah.. 



JL 



^ A/J BALB/c CBA/J C3H/H«J C87BI/6 OBA/2 

MOUSE STRAIN 

Fig. 2. Dexamethasone-induced suppression of NK 
activity of various mouse strains. The spleen cells of 
inbred mice of different genetic backgrounds were in- 
cubated 5 hr in control medium (D) or medium con- 
taining 10 ' M dexamethasone (■). After washing, the 
NK activity of each culture was determined against 
YAC-1 cells. Although tested at three different effec- 
tor to target cell ratios, only the mean percentage 
cytotoxicity (±SD) of the 100:1 cultures are shown in 
this representative experiment. 



DEXAMETHASONE-SUPPRESSED NK CULTURES 



149 



murine NK activity regardless of genetic 
background. 

The results of this study establish an in 
vitro model system for investigating the ac- 
tion of glucocorticoids on NK activity. The 
reduction of NK activity in mouse spleen 
cell cuhures after dexamethasone treatment 
is time dependent and dose dependent at 
pharmacologic concentrations. Since the 
proportion of cells possessing NK activity 
is thought to be less than 5% of the total 
splenic population, viability studies are of 
little value in determining whether DEX is 
toxic for NK effector cells (17). Nonethe- 
less, the viabilities of treated and control 
cultures are similar. Additional studies 
employing this in vitro system do not dem- 
onstrate the presence of suppressor cell ac- 
tivity after DEX treatment although NK 
activity is depressed (manuscript in prepa- 
ration). This would suggest that glucocor- 
ticoids suppress NK activity by acting di- 
rectly on the NK effector cell. 

In vitro suppression of NK activity by 
meaningful concentrations of glucocor- 
ticoids, to our knowledge, has not been 
previously demonstrated although Parillo 
and Fauci reported that human NK activity 
is decreased by the addition of very high 
concentrations of DEX (10"'^ Af and 10"^ Af , 
termed "pharmacologic" and **supraphar- 
macologic") to 18-hr cytotoxicity assays 
(9). Since these concentrations are known 
to produce nonspecific effects, the relevance 
of these results are questionable (18). 

We have obtained very similar in vitro 
glucocorticoid effects on human NK activ- 
ity (submitted for publication). Incubation 
of FicoU-Hypaque-enriched mononuclear 
cells of normal volunteers for 24 hr in 5 x 
10"^ M dexamethasone suppressed the NK 
activity to only 30% of the activity in cul- 
tures that had received control medium. 
Thus, the in vitro model system we describe 
here appears to closely simulate not only 
the in vivo glucocorticoid-induced suppres- 
sion of NK cytotoxicity in mice but also in 
humans. 

We gratefully acknowledge Dianne Strock for ex- 
pert technical assistance. Nan Rojas for assistance 
with the graphics, Allan Munck for critical reviews, 
and Pat Urban for typing this manuscript. This work 



was supported in part by USTHS Research Grants 
CA 17323 and AMD3535 and National Research Ser- 
vice Award CA09367. 

1. Herberman RB, Nunn MF, Lavrin DH. Natural 
cytotoxic reactivity of mouse lymphoid cells 
against syngeneic and allogeneic tumors. I. Dis- 
tribution of reactivity and specificity. Int J Cancer 
16:230-239, 1975. 

2. Kiessling RE, Klein E, Pross H, Wigzell H. 
"Natural" killer cells in the mouse. 11. Cytotoxic 
cells with specificity for mouse Moloney leukemia 
cells. Characteristics of the killer cell. Eur J Im- 
munol 5:117-121, 1975. 

3. Oehler JR, Lindsay LR, Nunn ME, Herberman 
RB. Natural cell-mediated cytotoxicity in rats. 1. 
Tissue and strain distribution, and demonstration 
of a membrane receptor for the Fc portion of IgG. 
Int J Cancer 21:204-209, 1978. 

4. Jondal M, Pross H. Surface markers on human B 
and T lymphocytes. VI. Cytotoxicity against cell 
lines as a functional marker for lymphocyte sub- 
populations. Int J Cancer 15:5%-605, 1975. 

5. Hansson M, Karre K, Kiessling R, Roder J, An- 
dersson B, Hftyry P. Natural NK-cell targets in 
the mouse thymus: Characteristics of the sensitive 
cell population. J Immunol 123:765-771, 1979. 

6. Gidlund M, Om A, Wigzell H, Senik A, Gresser I. 
Enhanced NK cell activity in mice injected with 
interferon and interferon inducers. Nature (Lon- 
don) 273:759-761, 1978. 

7. Herberman RB, Djeu JY, Ortaldo JR, Holden 
HT, West WH, Bonnard CD. Role of interferon in 
augmentation of natural and antibody-dependent 
cell-mediated cytotoxicity. Cancer Treatment Rep 
62:1893-18%, 1978. 

8. Senik A, Gresser 1, Maury C, Gidlund M, Om A, 
Wigzell H. Enhancement of mouse NK cells by 
interferon. Transplant Proc 11:993-9%. 1979. 

9. Parillo JE, Fauci AS. Comparison of the effector 
cells in human spontaneous cellular cytotoxicity 
and antibody-dependent cellular cytotoxicity: 
Differential sensitivity of effector cells to in vivo 
and in vitro corticosteroids. Scand J Immunol 
8:99- 107, 1978. 

10. Oshiani K, Gonda M, Sumiya M, Kano S. Effects 
of corticosteroids on natural killer cell activity in 
systemic lupus erythematosus. Clin Exp Immunol 
40:83-88, 1980. 

11. Shellam GR. Gross-virus-induced lymphoma in 
the rat. V. Natural cytotoxic cells are non-T-cells. 
Int J Cancer 19:225-235, 1977. 

12. Hochman PS, Cudkowicz G. Different sen- 
sitivities to hydrocortisone of natural killer cell 
activity and hybrid resistance to parental marrow 
grafts. J Immunol 119:2013-2015, 1977. 

13. Hochman PS, Cudkowicz G. Suppression of natu- 



150 



DEXAMETHASONE-SUPPRESSED NK CULTURES 



ral cytotoxicity by spleen cells of hydrocortisonc- 
treated mice. J Immunol 123:968-976. 1979. 

14. Lotzova E. Savary GA. Parallelism between the 
effect of cortisone acetate on hybrid resistance 
and natural killing. Exp Hematol 9:766-774. 
1981. 

15. Klein G, Klein E. Haughton GJ. Variation of an- 
tigenic characteristics between different mouse 
lymphomas induced by the Moloney virus. J Natl 
Cancer Inst 36:607-621. 1966. 

16. Brunner KT. Mauel J, Cerotinni JC. Chapuis B. 
Quantitative assay of the lytic action of immune 
lymphoid cells on ^ *Cr-labelled allogeneic target 



cells in vitro; inhibition by isoantibody and by 
drugs. Immunology 14:181-196. 1968. 

17. Roder JC. Kiessling R. Target -effector interac- 
tion in the natural killer cell system. I. Covariance 
and genetic control of cytolytic and target-cell- 
binding subpopulation in the mouse. Scand J Im- 
munol 8:135-144. 1978. 

18. Munck A. Steroid concentration and tissue integ- 
rity as factors determining the physiological sig- 
nificance of effects of adrenal steroids in vitro. 
Endocrinology 77:356-360. 1965. 

Received April 5. 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 151-157 (1982) 

Macromegakaryocytosis After Hydroxyurea^ (41490) 

SHIRLEY EBBE^ and ELIZABETH PHALEN 

Donner Laboratory, Lawrence Berkeley Laboratory, University of California, Berkeley, California 94720, 
and Department of Laboratory Medicine, University of California, San Francisco, California 94143 



Abstract. A single injection of hydroxyurea (OHU) produced transient megakaryo- 
cytopenia in mice. An increase in the average mean size of mature, stage III mega- 
karyocytes coincided with their depopulation. This was due to a selective reduction in 
numbers of smaller cells. In contrast, the macromegakaryocytosis of immunothrom- 
bocytopenia showed substantial increases in numbers of larger cells and reductions in 
smaller. Further reduction in numbers of smaller cells occurred when OHU was given to 
mice with immunothrombocytopenia, and the megakaryocytopenia was somewhat more 
severe than that produced by OHU in normal mice. OHU produced mild thrombocytopenia 
in normal mice and compromised recovery of the platelet count from inmiunothrom- 
bocytopenia. The most likely explanation for the increase in mean megakaryocyte size in the 
hypomegakaryocytic state produced by OHU is that the temporary imbalance between a low 
rate of influx and a normal rate of maturation produced a shift of the age distribution of the 
cells due to a deficiency of immature cells. 



Megakaryocyte size adjusts in response 
to perturbations of platelet count, demon- 
strating that it is one of the variables in the 
regulation of platelet production (1, 2). 
When megakaryocytopoiesis is stimulated 
or suppressed by perturbations of the 
platelet count, ploidy and size of mega- 
karyocytes are, respectively, increased or 
decreased (2, 3). Size is determined not 
only by ploidy but also by the level of 
maturation in the compartment of recog- 
nizable megakaryocytes (4). Within any 
maturation stage, megakaryocyte size is 
proportional to ploidy, and within any 
ploidy group, size is proportional to matu- 
rity (S, 6). Megakaryocyte maturation is 
subject to some degree of variability being 
accelerated in response to thrombocyto- 
penia (4) but remaining normal in trans- 
fusion-induced thrombocytosis (7). How- 
ever, instances in which changes in the 
age distribution of megakaryocytes alone 



> Supported, in part, by Grant R01-AM21355 from 
the National Institutes of Health and, in part, by the 
Office of Health and Environmental Research of the 
U.S. Department of Energy under Contract DE- 
AC03-76SP00098. 

' To whom all correspondence should be addressed. 



may have effected changes in mean cell size 
have not been described. 

In addition to the megakaryocyte size 
changes that can be produced by manipula- 
tion of the platelet count, mean megakaryo- 
cyte size may be increased in hypomega- 
karyocytic states in which thrombocytope- 
nia does not appear to be causative (8). The 
present studies were done to analyze the 
macromegakaryocytosis that accompanies 
the transient hypomegakaryocytic state fol- 
lowing a single injection of hydroxyurea 
(OHU). 

Materials and Methods. Female mice of 
the CF, strain (Charles River) were used at 
the age of 12-14 weeks. Each mouse was 
sampled only once; sequential studies were 
done with cohorts of mice. Blood for 
platelet counts was obtained by cardiac 
puncture under ether anesthesia and an- 
ticoagulated with dry KgEDTA: platelets 
were counted by phase microscopy (9). 
Cells were flushed out from each tibia with 
1 ml of 1% NajEDTA in saline for counts of 
nucleated cells by Coulter counter and 
megakaryocytes by microscopy with new 
methylene blue stain. 

Bone marrow smears were made from 
split femurs with a paint brush technique 



151 
0057-9721/%2J \«i\^\-QT%^\ .^^ 



152 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 



and stained with Wright's and Giemsa 
stains. Megakaryocytes were classified ac- 
cording to morphological criteria; all 
megakaryocytes were larger than myeloid 
and erythroid cells. Stage 1 megakaryocytes 
had basophilic cytoplasm without visible 
granules and a high nucleus/cytoplasm 
ratio. Stage 111 had azurophilic granules 
throughout the cytoplasm and a low 
nucleus/cytoplasm ratio. Stage II had in- 
termediate morphology. DifTerential counts 
were done to classify 100 megakaryocytes 
from each mouse as stage I, II, or III. Sizes 
of stages 1 and III megakaryocytes were 
determined by measuring the areas of im- 
ages enlarged from negatives of black and 
white photomicrographs; the areas were 
expressed as planimeter units. 

Anti-mouse platelet serum (APS) was 
produced in rabbits or guinea pigs; before 
use it was heat inactivated and absorbed 
three times with equal volumes of washed 
mouse red cells. It was injected ip in 0. 1 ml 
volumes after appropriate dilution with 
saline. 

OHU was freshly dissolved in saline be- 
fore injection. It was injected iv (tail vein) 
in a dose of 900 mg/kg body weight. 

Results. One day after a single injection 
of OHU, tibial megakaryocytes and total 
nucleated cells declined to about 80% of 
normal (Fig. I). Megakaryocytes decreased 
further to about half of normal on Days 2 
and 3 before beginning to recover on Day 4. 
Total cellularity was less severely di- 
minished and recovered about one day 
sooner than megakaryocytes. Cell counts 
returned to, but did not exceed, normal 
numbers during the week following admin- 
istration of OHU. 

A plot of average stage 111 megakaryo- 
cyte si/c (Fig. 2) was a mirror image of the 
plot of megakaryocyte numbers: size in- 
creased as number decreased, and both re- 
turned to normal at the same time. The 
maximum increase in mean size of stage 111 
megakaryocytes occurred on Days 2 and 3, 
and it was not preceded by an increase in 
mean si/c of stage I cells on Day 1. This 
finding confirmed earlier unpublished ob- 
servations in which rats were given 900 mg 
OHU/kg; for 3 days there was a gradual in- 



140 


o wtmnui ceiis 


- 


120 


k _ i 




CO 




iMs 


h 


- 


80 


H^'' 


rn 




- 


GO 


- Y_ 


f 


- 


40 


- r 




- 


20 
n 


1 1 1 1 i 1 . 


- 



2 4 6 8 

Ooys offer OHU 

Fig. 1. Numbers of tibial megakaryocytes and nu- 
cleated ceDs. expressed as percentage of control values, 
for 7 days after administration of OHU to mice. Mean ^ 
SEM for 33 controls is shown at time: other points 
represent 11-12 mice. 



crease in average size of stage III cells to 
about 140% of normal with no change in 
average size of stage I cells. 

To evaluate the changes in mean size of 
stage 111 cells, it was necessary to deter- 
mine the relative number of stage Ills at 
each sampling time after OHU. Differential 
counts of stages I, II, and III were done 
(Fig. 3, left); multiplication of percentage 
of each by the relative total number of 
megakaryocytes (Fig. 1: controls = 100) 
yielded the relative number of each stage in 
the marrow (Fig. 3, right). Megakaryocyte 
depletion occurred first in stage I cells and 
progressed to involve all three stages. 
Normal numbers of stage III cells were 
maintained through the first post-treatment 
day after which they decreased. On Days 2 
and 3, the 50% reduction in total mega- 
karyocytes was associated with a normal 
differential count, and therefore, equal re- 
duction in all stages. 

Size distribution curves for stage III 
megakaryocytes were prepared by deter- 
mining the percentage of the cells that were 
in each 50()-unit segment of the size range 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 



153 




4 6 

Ooyt after OHU 

Fig. 2. Mean megakaryocyte size (±SEM), ex- 
pressed as percentage of control values, after admin- 
istration of OHU to mice. Stage III control value rep- 
resents 1087 megakaryocytes from 33 mice; sub- 
sequent points, 376-386 megakaryocytes from 12 
mice. Stage I control value represents 474 megakaryo- 
cytes from 33 mice and, at one day, 118 megakaryo- 
cytes from 12 mice. 



(0-4000 units) at each sampling time. Each 
percentage was then multiplied by the rela- 
tive number of stage IIFs at the same sam- 
pling time to develop size distribution 
curves that were representative of the ac- 
tual number of stage III megakaryocytes 
and thus accounted for changing numbers 
of cells (Fig. 4). The 40% increase in mean 
size on Days 2 and 3 was due to a dispro- 
portionately great reduction in numbers of 
smaller megakaryocytes (<10(X) units) and 
retention of normal numbers of larger cells 
(>1250 units); intermediate cells were di- 
minished in proportion to the whole stage 
III population. Partial recovery on Day 4 
was due to influx of intermediate size cells 
rather than those at either extreme of the 
curve. 

Platelet counts after OHU are shown in 
Fig. 5; they were normal for 3 days, but 
declined to about 85% of normal on Days 4 
and S. 

The effects of OHU on APS-treated mice 
were similar when the OHU was given after 
1 day or after 4 days of APS-induced 
thrombocytopenia, so only the results of 
OHU after 1 day of thrombocytopenia are 




3 4 

Day« ofttr OHU 

Fig. 3. Left panel: differential count of stages I, II, 
and III megakaryocytes after administration of OHU 
to mice; 100 megakaryocytes from each of 33 controls 
and 12 mice at each time after OHU were classified. 
Results are means ± SEM. Right panel: relative number 
of each maturation stage was calculated by multiplying 
mean percentage (left panel) by relative mean total 
number of megakaryocytes (Fig. 1). 



presented. Because of variability in platelet 
and megakaryocyte counts, two experi- 
ments are presented individually; however, 
tibial cell counts were almost identical in all 
four experiments done with APS and OHU, 
thus indicating that there was a consistent 
response to the chemical in all experiments. 

Recovery from APS-induced thrombo- 
cytopenia was associated with variable de- 
grees of rebound thrombocytosis and in- 
creases in numbers of tibial megakaryo- 
cytes, but there was a consistent increase in 
mean size of stage III megakaryocytes (Fig. 
6). Administration of OHU retarded recov- 
ery of platelet counts and prevented re- 
bound thrombocytosis. Maximal reductions 
in megakaryocyte numbers occurred 2 days 
after administration of OHU and were 
about one-third to one-half of the concom- 
itant values in mice treated only with APS. 
Mean stage III megakaryocyte size was 
greater after combined treatment with APS 
and OHU than after APS alone, and the 
first appearance of a difference in size coin- 
cided with the reduction in number. 

Size distributions of stage III megakar- 
yocytes for the Day 2 samples were cor- 
rected for changing numbers of cells from 
the differential count of stages I, II, and 
III and total numbers of megakaryocytes 



154 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 




Fig. 4. Size distribution curves for stage III megakaryocytes from control mice and at intervals after 
administration of OHU. Abscissa expresses size in arbitrary units; ordinate is proportional to actual 
numbers of cells per tibia. Controls represent 1087 megakaryocytes from 33 mice; each curve after 
OHU represents 376-386 megakaryocytes from 12 mice. 



as described above. Size distribution curves 
which are, therefore, representative of the 
actual numbers of stage III megakaryocytes 
are presented for the two experiments in Fig. 
7. APS-induced thrombocytopenia alone 
was associated with reduced numbers of 
smaller megakaryocytes (< 1000 units), sub- 
stantial increases in larger ones (> 1250 units), 
and inconsistent appearance of a few cells 
larger than any normally found in the mar- 
row. OHU had the same effect on the size 



^ 140 
o 

1 


I 1 1 I 1 I I 


^120 


W\ 


L T 


6 lOOi 

t 80 

o 

a. 


r M 


1 


w/ 


> ^ 1 


60 


- 


n 


1 1 1 1 1 1 1 ' 



2 4 6 8 

Doys after OHU 

Fig. 5. Platelet counts, expressed as percentage of 
control, of 33 control mice and 11-12 mice at each 
sampling time after administration of OHU. Each 
point is the mean ± SEM. 




Fig. 6. Platelet counts, tibial megakaryocytes, and 
mean stage III megakaryocyte sizes in mice given APS 
on Days - I and with or without OHU on Day 0. Two 
experiments are shown. Each control is shown at 
time and represents eight mice: each other point repre- 
sents three to five mice. Each point is the mean ± SEM. 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 



155 



^ I I I — I 1 — t * I 




Fig. 7. Size distribution curves for stage III megakaryocytes from the same mice depicted in Fig. 6. 
Abscissa is size units; ordinate is proportional to actual numbers of cells per tibia. Controls represent 
289 and 160 megakaryocytes from seven and eight mice. APS treatment represents 174 and 94 
megakaryocytes from five mice in each experiment. APS + OHU treatment represents 175 and 88 
megakaryocytes from five mice in each experiment. Values for treated mice were obtained 2 days after 
the treatment. 



distribution of megakaryocytes in APS- 
treated mice as it had had in normals, i.e., 
retention of the largest cells, reduction in 
number of intermediate cells proportional 
to the reduction in the whole population, 
and disproportionately great reduction in 
numbers of smaller cells. 

Discussion. The present results help to 
explain how macromegakaryocytosis de- 
velops in the hypomegakaryocytic state 
produced by OHU (8). 

The number of stage I megakaryocytes 
was reduced as early as 4 hr after adminis- 
tration of OHU, indicating that the drug de- 
stroyed either stage I cells directly or pre- 
cursors that should have matured rapidly to 
become stage Ts. Longer al, (10) concluded 
that stage I megakaryocytes are actually a 
part of a population of smaller cells with 
lobed nuclei that could be identified his- 
tochemically as megakaryocytes. They did 
not find a cytocidal effect of OHU on this 
population 3 hr after its injection; rather, 
they found a 57% reduction in small precur- 
sor cells with round nuclei. Transit time for 
stage I rat and mouse megakaryocytes has 
been estimated to be 6-14 hr (11-13). 
Thus, the drop of about 50% at 4 hr could 
have been due to cessation of influx from a 
drug-sensitive compartment, but it seems 
more likely that some of the stage I cells 
were directly killed by OHU in these ex- 
periments. 



The deficiency of megakaryocytes was 
apparent in stages II and III 1 to 3 days after 
administration of OHU as the requisite time 
elapsed for the damaged populations to 
mature into those compartments. Persis- 
tence of low numbers of megakaryocytes 
for 3 days indicated that OHU destroyed 
cells in an important precursor compart- 
ment, probably a dividing 2N cell popula- 
tion. The delay in recovery could not be 
attributed to prolonged action of the drug, 
as Morse et aL (14) have shown that DNA 
synthesis resumes as soon as 2 hr after 900 
mg OHU/kg in the bone marrow of CF, 
mice. 

Depopulation of the stage III compart- 
ment was associated with an increase in the 
mean cell size, due to selective loss of 
smaller stage Ill's without an increase in 
numbers of larger cells. Thus it contrasted 
with the macromegakaryocytosis produced 
in response to immunothrombocytopenia in 
which larger megakaryocytes were in- 
creased in number at the apparent expense 
of smaller ones. The macromegakaryocyto- 
sis produced in response to peripheral throm- 
bocytopenia is known to be associated with 
increased ploidy (2, 3). Endoreduplication 
is completed and final ploidy determined 
before the cells leave stage I (11, 12, 15), 
and macrocytosis of stage I cells precedes 
that of stage III cells when stimulated by 
platelet depletion (1). Failure of stage Ts to 



156 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 



show increases in size after OHU shows 
that the mechanism responsible for stage III 
macrocytosis after OHU differs from that 
seen after thrombocytopenia. 

Megakaryocytes increase progressively 
in size as they mature from stage I through 
stage III (1); since stage III occupies 1-2 
days of the 2- to 3-day total transit time (12, 
13), it can be assumed that they continue to 
grow within this morphological group. Re- 
duced influx in the presence of a normal 
transit time would account for the presence 
of fewer immature, and smaller stage IIFs 
during the period when their number was 
dropping. The somewhat greater reduction 
in megakaryocytes produced by OHU in 
APS-stimulated mice than in normals could 
then be attributed to the shorter transit time 
that occurs in thrombocytopenic animals 
(4, 16). The uniqueness of the OHU-treated 
mice is the increase in mean megakaryocyte 
size while their number is dropping. The 
mere reduction in smaller megakaryocytes 
has been observed (17) in irradiated mice 
during periods of stable megakaryocyte 
counts in which, therefore, an imbalance 
between rate of influx and rate of matura- 
tion does not appear to be involved. 

Levin et aL (18) and Paulus et aL (19) 
have observed an inverse relationship be- 
tween numbers of cells and their ploidy in 
megakaryocyte colonies cultured from 
hemopoietic cells. In these in vitro systems, 
therefore, the more a precursor cell divides, 
the less it appears to endoreduplicate. If 
this principle applies//? vivo, the precursors 
of the smaller stage Urs might have divided 
more than the precursors of larger ones 
and, therefore, have been more heavily 
damaged by OHU. This notion, however, 
would imply that the larger cells that per- 
sisted after OHU would have a higher de- 
gree of polyploidy than those that disap- 
peared, and, as noted above, the failure of 
stage I cells to enlarge speaks against there 
being significant shifts in ploidy distribu- 
tion. 

OHU produced minor changes in platelet 
counts, probably because of the transient 
nature of the reduction in megakaryocytes. 
Considering that mouse platelets survive 
for about 4 days (20), the 15% drop in 
platelet count on Day 4 is consistent with a 



platelet production rate of about 50% for 
the precedent 24 hr, which, in turn, is con- 
sistent with the megakaryocyte number 
having been 50% of normal on Day 3. How- 
ever, platelet counts on Days 3 and 4 were 
somewhat greater than would have been 
expected if production had been reduced to 
50% for 2 days as might have been expected 
from the numbers of stage III megakaryo- 
cytes. It could be proposed that, even 
though total cell number was reduced on 
Day 2, those that remained represented a 
nearly normal complement of the most 
mature, platelet-forming, stage III cells. 

Recovery from immunothrombocytopenia 
was delayed, and rebound thrombocytosis 
was prevented, demonstrating a somewhat 
different effect for OHU than was seen when 
vincristine (VC) was administered to acutely 
or chronically thrombocytopenic rats (21). 
VC inhibited recovery from acute, but not 
chronic, thrombocytopenia whereas OHU 
inhibited recovery from both. The two 
drugs produced comparable reduction in 
megakaryocytes, so the reason for the dif- 
ference is not clear. However, a difference 
in ability to recover from acute immuno- 
thrombocytopenia has also been found be- 
tween genetically anemic mice of the W/W^ 
(22) and Sl/Sl^ (23) strains. Both strains 
are comparably hypomegakaryocytic, but 
the W/W responds normally to acute 
platelet depletion whereas the Sl/Sl'^ fails 
to develop rebound thrombocytosis. Nei- 
ther the drugs nor the genetic abnormali- 
ties interfered with the development of 
macromegakaryocytosis. These variable 
responses to thrombocytopoietic stimula- 
tion under different conditions indicate that 
platelet production can not be accurately 
predicted from just the number of mega- 
karyocytes and their mean size and that 
other factors must also be important. 



1. Ebbe S, Stohlman F Jr. Overcash J, Donovan J, 
Howard D. Megakaryocyte size in throm- 
bocytopenic and normal rats. Blood 32:383-392. 
1968. 

2. Odell TT, Murphy JR. Jackson CW. Stimulation 
of megakaryocytopoiesis by acute throm- 
bocytopenia in rats. Blood 48:765-775. 1976. 

3. Penington DC Olsen TE. Megakaryocytes in 
states of altered platelet production: Cell numbers. 



MACROMEGAKARYOCYTOSIS AFTER HYDROXYUREA 



157 



size and DNA content. Brit J Haematol 
18:447-463. 1970. 

4. Ebbc S, Stohlman F Jr, Donovan J, Overcash J. 
Megakaryocyte maturation rate in throm- 
bocytopenic rats. Blood 32:787-795, 1968. 

5. Odeil TT Jr, Jackson CW. Polyploidy and mat- 
uration of rat megakaryocytes. Blood 32:102- 
110, 1968. 

6. Ckicll TT Jr, Jackson CW, Friday TJ. Megakar- 
yocytopoiesis in rats with special reference to 
polyploidy. Blood 35:775-782, 1970. 

7. Ebbe S, Stohlman F Jr, Donovan J, Howard D. 
Megakaryocytopoiesis in the rat with transfusion- 
induced thrombocytosis. Proc Soc Exp Biol Med 
122:1053-1057, 1966. 

8. Ebbe S, Phalen £. Does autoregulation of 
megakaryocytopoiesis occur? Blood Cells 
5:123-138, 1979. 

9. Brecher G, Cronkite EP. Morphology and enu- 
meration of human blood platelets. J Appl Physiol 
3:365-377, 1950. 

10. Long MW, Williams N, Ebbe S. Immature 
megakaryocytes in the mouse: Physical charac- 
teristics, cell cycle status, and in vitro respon- 
siveness to thrombopoietic stimulatory factor. 
Blood 59:569-575, 1982. 

11. Feinendegen LE, Odartchenko N, Cottier H, 
Bond VP. Kinetics of megakaryocyte prolifera- 
tion. Proc Soc Exp Biol Med 111:177-182. 1962. 

12. Ebbe S, Stohlman F Jr. Megakaryocytopoiesis in 
the rat. Blood 26:20-35. 1965. 

13. Ebbe S. The megakaryocyte: maturation and 
self-renewal. In: Brinkhous KM, Shermer RW, 
Mostofi FK, eds. The Platelet. Baltimore, Wil- 
liams & Wilkins.pl, 1971. 

14. Morse BS, Rencricca NJ, Stohlman F Jr. The ef- 
fect of hydroxyurea on differentiated marrow 
erythroid precursors. Proc Soc Exp Biol Med 
130:986-989, 1969. 



15. Odell TT Jr. Jackson CW, Gosslee DG. Matura- 
tion of rat megakaryocytes studied by microspec- 
trophotometric measurement of DNA. Proc Soc 
Exp Biol Med 119:1194-1199, 1965. 

16. Odell TT Jr, Jackson CW, Friday TJ, Charsha DE. 
Effects of thrombocytopenia on megakaryocyto- 
poiesis. Brit J Haematol 17:90-101, 1969. 

17. Ebbe S, Phalen E, Threatte G, Adrados C. 
Megakaryocytopoiesis in irradiated, splenec- 
tomized mice. Exp Hematol 9:1020-1027, 1981. 

18. Levin J, Levin FC, Penington DG, Metcalf D. 
Measurement of ploidy distribution in 
megakaryocyte colonies obtained from culture: 
With studies of the effects of thrombocytopenia. 
Blood 57:287-297, 1981. 

19. Paulus JM, Prenant M, Maigne J, Henry- Amar 
M, Deschamps JF. Ploidization of megakaryocyte 
progenitors in vitro. In: Evatt BL, Levine RF, 
Williams NT, eds. Megakaryocyte Biology and 
Precursors: In Vitro Cloning and Cellular Prop- 
erties. New York, Elsevier/North-Holland, pl71, 
1981. 

20. Odell TT Jr, McDonald TP. Life span of mouse 
blood platelets. Proc Soc Exp Biol Med 
106:107-108, 1961. 

21. Ebbe S, Howard D, Phalen E, Stohlman F Jr. Ef- 
fects of vincristine on normal and stimulated 
megakaryocytopoiesis in the rat. Brit J Haematol 
29:593-603, 1975. 

22. Ebbe S, Phalen E. Regulation of megakaryocytes 
in W/W* mice. J Cell Physiol 96:73-80, 1978. 

23. Ebbe S, Phalen E, D'Amore P, Howard D. 
Megakaryocytic responses to thrombocytopenia 
and thrombocytosis in Sl/Sl** mice. Exp Hematol 
6:201-212, 1978. 



Received May 17, 1982. P.S.E.B.M. 1982, Vol. 171. 



f«OC££DfNOS OF THE SOCIETY FOft EXFEUMEBTTAL BIOLOCY AND MEMCIWE 171« 151- 1€3 U^CZ) 



Regional Uptake of [^]NorepinephrJne by the Canine Left Ventricle^ (41491) 

WILLIAM M, CHILIAN,* ROGER B. BOATWRIGHT, TETSURO SHOJI, and 

DOUGLAS M. GRIGGS, Jr.* 

Department of PhyMuAogy, University of Missouri School of MeScine, Cohtmbia, Missouri &52I2 



Abstract. The distribution of sympathetic oeunNis within the canine left ventricle was 
assessed hy measuring the myocardial uptake of pHtnorcpinephrine in anesthetized dogs. 
The nonneuronal uptake of pHlnorepinephrine was also assessed in a separate group of 
cocaine-treated animals. The left ventricle was systematicaOy divided into multiple sections. 
The pHjnorepinephrine was isolated in tissue homogenates, using alumina extraction, and 
quantified by liquid scintillation counting. The results revealed a nonuniform pattern of 
pH )norepinephrine uptake in the ventricular free waO, with a greater uptake in the base than 
in the apex. Differences in uptake between the anterior and posterior regions of the free waO, 
and between the free wall and the septum were not statistically significant. In the cocaine- 
treated animals uptake was approximately 20% of that in the control animals. Furthermore, 
there were no significant regional differences. These data suggest that in the canine left 
ventricle the sympathetic nerves are distributed nonuniformly between the base and apex, 
but otherwise the distribution is uniform. 



Although neural control of the heart is a 
topic of considerable current interest (1,2), 
very little information is available on the 
regional distribution of cardiac sympathetic 
neurons within the left ventricular myocar- 
dium. The sympathetic nerve supply to the 
left ventricle has been shown to cross from 
the pulmonary artery-aortic region to the 
ventricle and to course in the epicardium in 
a base to apex direction before branching 
and innervating the deeper layers of the 
myocardium (3-5). In previous studies 
concerned with regional innervation of the 
canine left ventricle, Angelakos (6) found a 
difference in norepinephrine content be- 
tween the base and the apex, and Dahl- 
strOm et al. (7) obtained histological evi- 



' 1 his work was supported by Grants ROl-HLl 1876 
and T-32-H1.07094 from U.S. Public Health Service. 
This paper was presented in part at the annual meet- 
ings of the Federation of American Societies for Ex- 
perimental Biology in Anaheim, California, April 1980. 
This work was part of Dr. W. M. Chilian's doctoral 
dissertation. 

' l>rcscnl address: Dr. W. M. Chilian, Department of 
Internal Medicine and Cardiovascular Center, College 
of Medicine. University of Iowa, Iowa City, Iowa 
.•^2242. 

^ To whom requests for reprints should be sent. 



dence of nonuniform transmural sympa- 
thetic innervation. 

The purpose of the present study was to 
characterize further the regional sympa- 
thetic innervation of the canine left ventri- 
cle. To accomplish this we utilized the 
[^H]norepinephrine uptake method of Kay e 
and Tyce (8), in which [^H]norepinephrine 
uptake has been shown to be a reliable 
index of cardiac sympathetic innervation. 
Results were obtained in both normal and 
cocaine-treated animals to distinguish be- 
tween the neuronal and nonneuronal uptake 
of pH]norepinephrine in different regions 
of the ventricle. 

Methods. The experiments were per- 
formed on 13 male mongrel dogs (20-40 kg) 
who had been maintained on a nourishing 
diet for at least 30 days. The animals were 
fasted overnight, premedicated with mor- 
phine sulfate (2.5 mg/kg, sc), and anes- 
thetized 45-60 min later with a-chloralose 
(100 mg/kg, iv). Supplemental doses of a- 
chloralose were given as required prior to 
the administration of [^H]norepinephrine, 
but additional anesthetic was avoided 
thereafter. The trachea was intubated with 
a cuffed endotracheal tube and the animal 
was ventilated with a Harvard respirator 
(Model 607). Supplemental oxygen was 



158 

'27/8yiOOOI5H'06$0l,00/0 

V 1^2 by the Society for Hxperimetital Biology and Medicine. 



MYOCARDIAL [>H]NOREPINEPHRINE UPTAKE 



159 



added to the inspired air to maintain a nor- 
mal arterial oxygen tension as determined 
on an Instrumentation Laboratory blood- 
gas analyzer (113-Sl). The right femoral 
artery and vein were isolated and catheter- 
ized, with the catheters advanced to the 
aortic arch and inferior vena cava. Arterial 
pressure was monitored with a Statham 
pressure transducer (P23Db) and an Elec- 
tronics for Medicine (Model Dr-8) oscillo- 
graph. A left thoracotomy was performed 
through the fifth intercostal space, the 
pericardium incised, and the heart exposed. 
Gauze, damped in saline (0.9%), was placed 
on the left ventricular epicardium to pre- 
vent the epicardial surface from drying. 
Heparin (500 U/kg) was administered and a 
blood sample was taken for determination 
of arterial PO2, PCO2, and pH. Blood sam- 
ples were analyzed immediately, and if re- 
quired, ventilation was adjusted to maintain 
blood gases within normal ranges and 
sodium bicarbonate was given to maintain 
arterial pH above 7.30. 

In eight animals l-[7,8-^H]norepinephrine 
(Amersham, 30-40 Ci/mmole) was admin- 
istered (2.5 /xCi/kg, iv) as described by 
Kaye and Tyce (8). The protocol involved 
the inftision of the pH]norepinephrine over 
a 5-min period, followed by a 20-min incu- 
bation period, which enabled the labeled 
norepinephrine to become incorporated 
into the neuronal pools and to clear the ex- 
tracellular space. At the end of the 20 min 
the heart was rapidly excised and placed in 
chilled saline (0°). In five additional ani- 
mals, cocaine was administered (iv) to 
block neuronal uptake, 330 /xg/kg/min for 15 
min prior to the pH]norepinephrine infu- 
sion and 66 /xg/kg/min during the infusion 
and incubation periods (9). 

Using a grid system, the left ventricle 
was divided into 33 sections in a cold room 
(25 transmural free wall sections and 8 
septal sections) (Fig. 1). By utilizing the 
papillary muscles as landmarks it was pos- 
sible to section the free wall into five 
*'rows" from base to apex and five **col- 
umns" from anterior to posterior. All tissue 
samples were minced, weighed, and placed 
into tubes containing 0.4 N perchloric acid 
{(f). The minced sample was homogenized 



with two 10- sec bursts of a polytron. The 
homogenate was centrifuged at 10,00Qg for 
20 min, the supernatant was decanted and 
frozen for later analysis. Two milliliters of 
the supernatant was placed into a tube 
containing 150 mg of alumina, which had 
been prepared according to the method of 
Anton and Say re (10), and 3 ml of 2 Af 
Tris-0.5 M EDTA buffer, pH 8.6 (at this 
pH norepinephrine is bound to the alumina, 
whereas the o- methylated metabolites are 
not). This solution was vortexed thor- 
oughly, and the supernatant was aspirated. 
The alumina was washed three times with 
distilled water (2-3 ml) and the norepi- 
nephrine was eluted from the alumina by 
addition of 1 ml of 0.05 N perchloric acid 
and thorough vortexing. The acid was 
pipetted into a scintillation tube containing 
10 ml of a toluene-based cocktail (3a20, Re- 
search Products International) and 3 ml of a 
detergent (Triton X-100 scintillation grade. 
Research Products International). Samples 
were counted for 10 min in a Packard Tri- 
Carb liquid scintillation spectrometer. 
Quenching standards were used to calculate 
efficiency which averaged 44%. pH]Nor- 
epinephrine standards were used to calcu- 
late the percentage yield of the alumina ex- 
traction procedures, which averaged 41%. 
All sample counts were corrected for tissue 
weight, efficiency, percentage yield, and 
counting time, yielding results expressed as 
disintegrations per minute per gram (dpm/g). 
When duplicate pH]norepinephrine uptake 
values were compared, the coefficient of 
variation (r^) was 0.95. 

Differences among the means of the 33 
sections for each of the two groups (control 
and cocaine-treated) were analyzed by 
analysis of variance. If the F value was sig- 
nificant (P < 0.05), a further statistical 
analysis was performed on the 25 free wall 
sections as follows: the data for individual 
sections were combined into 5 horizontal 
**rows'* for making comparisons between 
the base and apex and parallel intermediate 
regions and into 5 vertical * 'columns" for 
making comparisons between the anterior 
and posterior regions and parallel inter- 
mediate regions, using analysis of variance 
and multiple comparison testing. 



160 



MYOCARDIAL [»H)NOREPINEPHRINE UPTAKE 



Results. The results obtained in the eight 
control animals are shown in Figs. 1-3. 
Shown in Fig. 1 are the pH]norepinephrine 
uptake data for all 33 tissue sections. 
Analysis of variance of the 33 sections was 
significant {P < 0.05). Since this indicated 
that the uptake of pH]norepinephrine was 
nonuniform, further statistical testing 
was performed. Shown in Fig. 2 are the 
pHJnorepinephrine uptake data for the 
larger regions of the left ventricular free 
wall obtained by combining sections into 
*'rows." Analysis of variance was signifi- 
cant {P < 0.05), indicating that pH]norepi- 
nephrine uptake was nonuniform in a base 
to apex direction. Further analysis revealed 
that uptake was significantly greater in the 
two most basilar ''rows'' than in the apical 
''row." Uptake in the two intermediate 
"rows" adjacent to the apex was not sig- 
nificantly different from that in the apical 
"row" or the two most basilar "rows." 




Septum 




Septum 



Fig. I. [^H]Norepinephrine uptake in 25 regions of 
the left ventricular free wall and 8 regions of the inter- 
ventricular septum obtained by systematically sec- 
tioning the ventricle into 33 tissue samples. The values 
shown, multiplied by I(P, represent disintegrations per 
minute per gram. They are the means and standard 
errors for eight control animals. 



Fig. 2. Myocardial ['H]norepinephrine uptake data 
(dpm/g) as analyzed in a base to apex direction. Values 
are mean ± SEM. 

Shown in Fig. 3 are the pH]norepineph- 
rine uptake data for the larger regions of the 
left ventricular free wall obtained by com- 
bining sections into "columns." Analysis 
of variance was insignificant (P > 0.1), in- 
dicating that pH]norepinephrine uptake 
was uniform in an anterior to posterior di- 
rection. Analysis of variance of the eight 
sections of the septum was insignificant (P 
> 0.05), indicating that pH]norepinephrine 
uptake was uniform in the septum. 

The average pH]norepinephrine uptake 
value for the entire left ventricular free wall 
was 20,000 ± 1200 dpm/g, whereas that for 
the septum was 16,000 ± 2000 dpm/g. The 
difference was not statistically significant 
{P >0.1). 

Results obtained in the cocaine-treated 
animals are shown in Fig. 4. Depicted are 
the pH]norepinephrine uptake data for all 
33 tissue sections. Analysis of variance of 
the 33 sections was insignificant {P > 0.1), 
indicating that nonneuronal pHJnorepi- 
nephrine uptake in the left ventricular free 
wall and septum was uniform. The average 



MYOCARDIAL [»H)NOREPINEPHRINE UPTAKE 



161 




Fig. 3. Myocardial pH]norepinephrine uptake data 
(dpm/g) as analyzed in an anterior to posterior ven- 
tricular wall direction. Values are mean ± SEM. 

pH]norepinephrine uptake value for the 
entire left ventricular free wall was 4000 ± 
500 dpm/g, whereas that for the septum was 
3500 ± 500 dpm/g. pH]Norepinephrine 
uptake for the free wall and septum in the 
cocaine-treated animals was significantly 
less than that in control animals, with the 
values averaging approximately 20% of 
those obtained in the control animals. 

Discussion. The major findings of this 
study are: (i) that a regional difference in 
neuronal pH]norepinephrine uptake was 
observed between the base and apex of the 
ventricular free wall, with the uptake being 
approximately 20% greater in the base than 
in the apex, and (ii) nonneuronal |^H]nor- 
epinephrine uptake is uniform throughout 
the left ventricle, and it amounts to ap- 
proximately 20% of that taken up by the 
normal ventricle. 

The method of pH]norepinephrine up- 
take to assess cardiac sympathetic efferent 
innervation has been extensively utilized by 
several laboratories (8, 11, 16). pH]Norepi- 
nephrine uptake was reported to be a more 




Septum 





4.7 
±0.7 


3.8 
±0.8 


3.1 
±0.9 


3.6 
iO.5 


3.2 
±0.5 


3.1 
iO.5 


\ 


3.0 
±0.6 


3.6 
±0.5 


/ 



Fig. 4. pHJNorepinephrine uptake portrayed as in 
Fig. 1, but for five cocaine-treated animals. 

reliable indicator of sympathetic innerva- 
tion than norepinephrine content of the 
heart (8). Alumina extraction of norepi- 
nephrine eliminates methylated metabo- 
lites, which have the tritium label, but not 
the deaminated metabolites. However, in 
the heart the deaminated metabolites are 
reported to constitute only 1-2% of the re- 
covered label after tracer norepinephrine 
uptake studies (17). The recovery of nor- 
epinephrine in this study was 41%, which is 
lower than that usually reported in the liter- 
ature. However, the recovery was consis- 
tent among the experiments and the coeffi- 
cient of variation of the procedure was 0.95. 
When the absolute uptake values were 
normalized for the dose of tracer norepi- 
nephrine, and compared to those of Kaye 
and Tyce (8), also normalized for the dose 
of tracer norepinephrine, the values were 
virtually identical. Thus, we believe our re- 
sults reflect the uptake of pH]norepineph- 
rine and the pattern of sympathetic inner- 
vation. 

In the canine left ventricle, the sym- 
pathetic innervation has been termed 



162 



MYOCARDIAL [>H]NOREPlNEPHRINE UPTAKE 



*'patchy" (7) and two earlier studies 
suggested that innervation is greater in the 
base than in the apex (6, 7). Evidence that 
the catecholamine content of the base is 
higher than that of the apex has also been 
provided in a more recent study (8). How- 
ever, in that study, no difference of pH]- 
norepinephrine uptake was found between 
the base and apex. The reason for the 
difference in findings between that study 
and the present study is not clear. One pos- 
sible explanation is a more discrete separa- 
tion of the ventricular free wall into apical 
and basilar regions in the present study. 
Only the pH]norepinephrine uptake value 
for the apical "row" was significantly dif- 
ferent from that for the most basilar 
*'rows," whereas the value for the inter- 
mediate ''row'* immediately adjacent to the 
apical ''row" was not significantly different 
from the two most basilar ''rows." Others 
have demonstrated that the sympathetic 
efferent innervation courses in the epicar- 
dium from the anterior basilar region to the 
apical and posterior regions (3). In addition, 
it has been shown that when the canine 
heart is surgically denervated, reinnerva- 
tion returns in a base to apex sequence (11). 
The present results support the notion of a 
greater sympathetic innervation in the base 
than in the apex. The pH]norepinephrine 
uptake was approximately 20% greater in 
the basilar region than in the apical region. 

One reason for regional variations in 
sympathetic innervation of the ventricle 
could be an association of the sympathetic 
nerves with the coronary vasculature. It 
has been shown that sympathetic nerves 
travel in the adventitia of the coronary con- 
ductance vessels before innervation of the 
myocardium (12, 13). We have shown in 
another study (14) that by applying phenol 
to selective sites on the myocardium and 
epicardial conductance vessels it is possible 
to produce a regional sympathectomy of the 
left ventricle. 

In cocaine-treated animals the pH]nor- 
epinephrine uptake was uniform through- 
out the ventricle. These results favor the 
conclusion that the greater basilar pH]- 
norepinephrine uptake in the normal an- 
imal was due to a regional difference in 



sympathetic innervation rather than to a re- 
gional difference in myocardial blood flow 
or some other experimental variable. More 
direct evidence against a regional difference 
in blood flow has also been obtained in 
other studies on this animal preparation, 
using the microsphere method to measure 
blood flow. The nonneuronal uptake sites 
consist of myocytes, vascular smooth mus- 
cle, connective tissue and fibroblasts (15), 
which are distributed homogeneously in the 
myocardium. 

The physiological significance of a 
nonuniform distribution of sympathetic 
nerves between the base and apex of the 
ventricular wall is not elucidated by the 
present study. Others have demonstrated 
localized changes in myocardial function by 
stimulating discrete branches of the cardiac 
sympathetic nerve supply (2) which sug- 
gests that overall ventricular function may 
be modulated by regional differences in 
cardiac sympathetic nerve function. 

In conclusion, the present study provides 
further evidence of a nonuniform distribu- 
tion of sympathetic neurons between the 
base and apex of the canine left ventricle, 
with a greater density of neurons in the base 
than in the apex. The regional distribution 
of sympathetic neurons otherwise appears 
to be uniform in the left ventricular free 
wall and septum. The study also indicates 
that the nonneuronal uptake of norepineph- 
rine is uniform in the left ventricle, and the 
nonneuronal uptake constitutes approxi- 
mately 20% of labeled norepinephrine up- 
take. 



1. Martins JB, Zipes DP. Effects of sympathetic and 
vagal nerves on recovery properties of the en- 
docardium and epicardium of the canine left ven- 
tricle. Circ Res 46:100-110, 1980. 

2. Randall W. Neural Control of the Heart. New 
York, Oxford Univ Press, 1977. 

3. Kaye MP. Brynjolfsson GG. Geis WP. Chemical 
epicardiectomy. Cardiologia 53:139-149, 1968. 

4. Martins JB, Zipes DP. Epicardial phenol inter- 
rupts refractory period responses to sympathetic 
but not vagal stimulation in canine left ventricular 
epicardium and endocardium. Circ Res 47:33-40, 
1980. 

5. Randall WC. Szentivanyi M, Pace JB, Wechsler 



MYOCARDIAL [»H)NOREPINEPHRINE UPTAKE 



163 



JS, Kayc MP. Patterns of sympathetic nerve pro- 
jections onto the canine heart. Circ Res 
22:315-323, 1968. 

6. Angelakos ET. Regional distribution of catechol- 
amines in the dog heart. Circ Res 16:39-44, 1965. 

7. Dahlstrom A, Fuxe K, Mya-Tu M. Zetterstrom 
BEM. Observations on adrenergic innervation of 
dog heart. Amer J Physiol 209:689-692, 1965. 

8. Kayc MP, Tyce GM. Norepinephrine uptake as an 
indicator of cardiac reinnervation in dogs. Amer J 
Physiol: Heart Circ Physiol 4:H289-H294, 1978. 

9. Levy MN, Blattberg B. The influence of cocaine 
and desipramine on the cardiac responses to ex- 
ogenous and endogenous norepinephrine. Eur J 
Pharmacol 48:37-49, 1979. 

10. Anton AH. Sayer DF. A study of the factors af- 
fecting the aluminum oxide-trihydroxyindole pro- 
cedure for the analysis of catecholamines. J 
Pharmacol Exp Ther 138:360-375, 1962. 

11. Kaye MP, Randall WC, Geis WP, Priola DV. 
Chronology and mode of reinnervation of the sur- 
gically denervated canine heart: Functional and 
chemical correlates. Amer J Physiol: Heart Circ 
Physiol 2:H431-H437, 1977. 

12. Denn MJ, Stone HL. Autonomic innervation of 
dog coronary arteries. J AppI Physiol 41:30-35, 
1975. 



13. McKibben JS, Getty R. A comparative mor- 
phologic study of the cardiac innervation in 
domestic animals. I. The canine. Amer J Anat 
122:533-544, 1968. 

14. Chilian WM, Boatwright RB, Shoji T, Griggs Jr 
DM. Evidence against significant resting sympa- 
thetic coronary vasoconstrictor tone in the con- 
scious dog. Circ Res 49:866-875. 1981. 

15. Trendelenburg U. The extraneuronal uptake and 
metabolism of catecholamines in the heart. In: 
Paton DM, ed. The Mechanism of Neuronal and 
Extraneuronal Transport of Catecholamines. New 
York, Raven, 1976. 

16. Coglianese C, Randall WC, Filkins JP. The rela- 
tive density of sympathetic nerve terminals in the 
canine right atrium. Proc Soc Exp Biol Med 
154:127-130, 1977. 

17. Kaye MP, Wills DJ, Tyce GM. Nerve growth 
factor-enhanced reinnervation of surgically de- 
nervated canine heart. Amer J Physiol: Heart Circ 
Physiol 5:H624-H628, 1979. 

18. Herting G, AxeU-od J, Whitby LG. Effect of drugs 
on the uptake and metabolism of ^H-norepineph- 
rine. J Pharmacol Exp Ther 134:146-153, 1961. 

Received December 28. 1981. P.S.E.B.M. 1982. Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 164-167 (1982) 



Stimulation of Mammary Tumorigenesis and Suppression of Uterine Adenomyosis 
by Temporary Inhibition of Pituitary Prolactin Secretion during Youth in 

Mice' (41492) 

HIROSHI NAGASAWA* and TAKAO MORI* 

Experimental Animal Research Laboratory, Meiji University, Tama-ku, Kawasaki, Kanagawa 214, and 
*Zoological Institute, Faculty of Science, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan 



Abstract. The effects of daily subcutaneous injections of 0.2 mg CB-154 (bromocrip- 
tine-mesilate), a potent suppressor of pituitary prolactin secretion, between 4 and 11 weeks 
of age on the occurrence of spontaneous mammary tumors and adenomyosis, a hyperplasia 
of endometrial tissue, were studied in the SHN strain of virgin mice. While there was little 
difference in mammary tumor incidence between experimental and control mice until 9 
months of age, mammary tumor incidence in the experimental mice given CB-154 was 
significantly enhanced and surpassed that in the control after 10 months; 25 (53.2%), 31 
(66.0%), and 36 (75.5%) of 47 experimental mice and 20 (27.5%), 23 (33.3%), and 34 (49.3%) 
of 69 control mice developed mammary tumors at 10, 1 1, and 12 months of age, respectively. 
In contrast, no adenomyosis appeared in 39 experimental mice at necropsy at 12 months of 
age, while 15 (46.9%) of 32 control mice developed it. Furthermore, five mice (15.6%) of the 
control had numerous subserosal nodules, an advanced state of adenomyosis. No significant 
correlation was observed in the control mice between the occurrence of mammary tumors 
and that of adenomyosis. 



Mammary gland DNA synthesis, which 
is primarily controlled, in part, by prolactin 
(1, 2), is a limiting factor for mammary 
tumorigenesis (3, 4). DMBA (7,12-di- 
methy lbenz[a ]anthracene)-induced mam- 
mary tumorigenesis was much more marked 
in rats given DMBA at proestrus, when 
both circulating prolactin and mammary 
gland DNA synthesis were high, than in 
rats given DMBA at diestrus, when pro- 
lactin and DNA synthesis were low (5). 
Suppression by CB-154 of the high prolac- 
tin at proestrus resulted in the decline of 
mammary gland DNA synthesis and the in- 
hibition of DMBA-induced mammary 
tumorigenesis. By contrast, the single pro- 
lactin injection elevated mammary gland 
DNA synthesis and stimulated DMBA- 
induced mammary tumorigenesis (5). More- 
over, temporary suppression of pituitary 
prolactin secretion during youth pro- 



' This work was supported partly by the grant-in-aid 
for Cancer Research from the Ministry of Education. 
Science and Culture. Japan (No. 56010048). 

^ To whom all correspondence should be addressed. 



tected markedly spontaneous mammary 
tumor development in rats (6, 7). Mammary 
gland DNA synthesis in this species is high 
only during youth with a peak around 7 
weeks of age decreasing thereafter with in- 
creasing age (5, 8). Russo and Russo (9) 
also found that mammary gland of 50-day- 
old virgin rats contained much higher 
number and labeling index of terminal ducts 
and end buds, from which DMBA-induced 
mammary tumors arise, than the glands of 
180-day-old virgin and multiparous rats. 
Thus, prolactin suppression after the peak 
of mammary gland DNA synthesis (11-18 
weeks of age) had little effect on prophylax- 
is of mammary tumors at advanced ages (6, 7). 
On the other hand, mammary gland DNA 
synthesis in SHN mice, a high mammary 
tumor strain, changes little with age and 
much higher than that in rats (10). Although 
chronic prolactin suppression was found to 
inhibit markedly spontaneous mammary 
tumor development in C3H mice (11-13), 
no data are available in this species on the 
effects of temporary suppression of prolac- 
tin secretion during youth on mammary 
tumorigenesis. The effects of short-term 



164 
0037-9727/82/IOOI64-04$Ol. 00/0 
Copyright (c=) I9ft7 h^ /A* Society for Experimental Biology and Medicine. 
^^ rights rer 



MAMMARY TUMOR AND ADENOMYOSIS IN MICE 



165 



and long-term suppression of prolactin se- 
cretion on mammary tumors are two differ- 
ent problems. 

It has recently been observed that SHN 
virgin mice develop spontaneously adeno- 
myosis, hyperplasia of endometrial tissue, 
both glandular and stromal components in 
the myometrium (14, 15). The development 
of adenomyosis is strongly prolactin de- 
pendent in the presence of ovarian hor- 
mones; ectopic pituitary grafts have been 
reported to enhance the appearance of the 
condition in intact animals (14, 15), how- 
ever, the mechanism of its development is 
little understood. 

The primary objective of this experiment 
was to study the effects of temporary sup- 
pression of prolactin secretion during youth 
on the development of spontaneous mam- 
mary tumors and adenomyosis at advanced 
ages in SHN mice. 

Materials and Methods. Mice. SHN 
strain of mice maintained by brother x sis- 
ter mating were used at the 45th generation. 
One of the characteristics of this strain is 
the high and early development of mam- 
mary tumors; the final incidence and onset 
age of tumors in virgin mice are 100% and 
8.9 months, respectively (16). They also 
develop adenomyosis after 7 months of age 
with the incidence of about 50% at 12 
months (14, 15). Throughout the experi- 
ment, five or six mice each were kept in 
Teflon cages (15 x 18 x 13 cm) with wood 
shavings, maintained in an air-conditioned 
(24 ± O.S^'C and 65-70% relative humidity) 
and artificially illuminated (14 hr of light 
from 5:00 am to 7:00 pm) animal room and 
provided with a commercial diet and tap 
ViSLttr ad libitum. 

CB'I54 treatment. A daily dose of 0.2 mg 
CB-154 (bromocriptine-mesilate: Sandoz 
Ltd., Basel, Switzerland), a potent sup- 
pressor of pituitary prolactin release 
(17-19), suspended in 0.05 ml olive oil with 
the Teflon glass homogenizer was injected 
subcutaneously each into 47 experimental 
mice between 4 and 11 weeks of age. 
Sixty-nine control mice received vehicle 
only. 

Mammary tumorigenesis. Each mouse 
was checked for palpable mammary tumors 



every 7 days between 3 and 12 months 
of age. 

Adenomyosis. All mice were killed by 
cervical dislocation at 12 months of age. 
Uteri were fixed in Bouin's solution, em- 
bedded in paraffin, sectioned at 7 /xm, and 
stained with hematoxylin -eosin for his- 
tological determination of adenomyosis 
(14, 15). 

Statistics. Statistical evaluation of mam- 
mary tumorigenesis was performed using 
the multiple classification method (two-way 
analysis of variance) (20). By this method, 
the statistical sequence of the incidence and 
onset age of mammary tumors could be de- 
termined simultaneously. The difference in 
the incidence of adenomyosis or subserosal 
nodules was evaluated by x^ test. 

Results. The results of mammary tumor 
development are presented in Fig. 1. In 
both experimental and control groups, the 
cumulative incidence of mammary tumors 
increased with age. However, the increas- 
ing rate was more marked in the experi- 
mental mice given CB-154 than in the con- 
trol, especially after 10 months. The 
number (and percentage) of mice with 
tumors at 10, 11, and 12 months were 25 
(53.2%), 31 (66.0%), and 36 (75.5%) of 47 
experimental mice and 20 (27.5%), 23 
(33.3%), and 34 (49.3%) of 69 control mice, 
respectively. Therefore, mammary tumori- 
genesis in the experimental group was sig- 
nificantly higher than that in the control 



'. 



alll 



Aq* imomhsJ 



Fig. 1. Cumulative incidence of mammary tumors 
in experimental mice receiving CB-154 between 4 and 
1 1 weeks of age (■) and the controls given vehicle 
only (D). 



166 



MAMMARY TUMOR AND ADENOMYOSIS IN MICE 



when evaluated by analysis of variance (P 
< 0.05). 

In contrast, none of 39 experimental mice 
developed adenomyosis by 12 months of 
age, whereas in the controls, 15 or 46.9% of 
32 mice developed it and five mice (15.6%) 
further had several subserosal nodules. The 
differences between experimental and con- 
trol groups in the incidences of these 
pathological endometrial states were statis- 
tically highly significant (P < 0.01) (Table I). 

No significant relationship was seen in 
the control mice between the occurrence of 
mammary tumors and that of adenomyosis; 
6 out of 10 mice bearing mammary tumors 
(60%) and 9 out of 22 mice with no tumors 
(41%) developed adenomyosis. 

The numbers of mice which died without 
tumors during the experiment were two and 
three in the experimental and control 
groups, respectively. 

The body weight at the beginning of CB- 
154 injection (4 weeks of age) was 17.6 ± 
0.2 (SE) g (pooled data of experimental and 
control groups) and the weights at the end 
of injection (11 weeks of age) were 26.9 ± 
0.5 and 27.7 ± 0.4 g in the experimental and 
control groups, respectively, showing no 
difference between groups. 

Discussion. This study shows that mice 
treated with CB-154 at an early age en- 
hanced mammary tumorigenesis when 
compared with the control. Two major 
factors could be responsible for the en- 
hanced mammary tumorigenesis seen in 
this study — pituitary and ovarian secretion 
of mammotropic hormones and mammary 
gland susceptibility to these hormones. Pi- 
tuitary prolactin and ovarian estrogen and 
progesterone are prerequisite for the devel- 



opment of mammary tumors in mice, while 
established mammary tumors in many 
strains do not require hormonal support 
and, therefore, are autonomous (21). The 
enhancement of mammotropic hormone se- 
cretion as a result of terminating CB-154 
treatment is unlikely, since normal repro- 
ductivity was observed in studies involving 
chronic administration of CB-154 (7, 22) 
and prolactin secretion at an advanced age 
was not altered by the treatment (7). 

Mammary gland susceptibility to mam- 
motropic hormones may be more important 
for normal and neoplastic mammary gland 
development than the secretion of mam- 
motropic hormones (2). It has been re- 
ported that mammary glands exposed to 
abnormal hormonal conditions during early 
development show an increased suscepti- 
bility to mammotropic hormones (2). The 
enhanced mammary tumorigenesis in this 
study may be ascribed to the long-term ef- 
fects of stimulated mammary gland suscep- 
tibility. 

In any case, the present results suggest 
that the temporary inhibition of mammary 
gland DNA synthesis through prolactin 
suppression (6, 7) is not effective in pre- 
venting mammary tumorigenesis in species 
where mammary gland DNA synthesis 
continues at a high rate throughout the 
lifetimes. 

The occurrence of adenomyosis at an ad- 
vanced age (12 months) was completely 
eliminated by the temporary CB-154 treat- 
ment. While no information is available on 
the mechanism of adenomyosis develop- 
ment in mice except for its high prolactin 
dependency (14, 15), this study has demon- 
strated that there is a critical period for its 



TABLE I. Incidence of Adenomyosis at 12 Months of Age in Experimental and Control Mice 





No. of 
mice 


No. (and %) of 


mice 


with 


Group" 


Adenomyosis 




Subserosal 
nodules 


Experimental 
Control 


39 
32 


0(0)* 
15 (46.9)'^ 




O(O)'' 
5(15.6r 



° Experimental and control mice received daily subcutaneous injections of CB-154 (0.2 mg) and vehicle only 
for 7 weeks between 4 and II weeks of age. respectively. 
hlc: die: P < 0.01. 



MAMMARY TUMOR AND ADENOMYOSIS IN MICE 



167 



expression similar to that observed in rats 
for spontaneous mammary tumors (6, 7). 
Normal uterine growth is dependent upon 
estrogen and progesterone from the ovary 
and the susceptibility of uterine cells to 
these hormones changes little with age (23). 
However, exposure of uterus to abnormal 
hormonal conditions during the early de- 
velopmental stages often induces a decrease 
in the susceptibility to estrogen (24). Thus, 
the present findings may indicate a con- 
tinuous decline in uterine susceptibility to 
prolactin and ovarian steroid hormones as a 
result of CB-1S4 injection during youth and 
this could contribute to the complete elimi- 
nation of adenomyosis at advanced ages. 

We thank Professor R. R. Gala, Department of 
Physiology, Wayne State University School of 
Medicine, Detroit, Michigan, for his reading of the 
manuscript and invaluable comments and Professor E. 
FlOckiger, Sandoz Ltd., Basel, Switzerland, for his 
kind donation of CB-154. 

1. Nagasawa H. Prolactin: Its role in the develop- 
ment of mammary tumours. Med Hypoth 
5:1117-1121. 1979. 

2. Nagasawa H. Hormones and experimental mam- 
mary tumorigenesis. In: Nagasawa H and Abe K, 
eds. Hormone Related Tumors. Tokyo/Berlin, 
Japan Sci Soc Press/Springer- Verlag, pp 137- 163, 
1981. 

3. Nagasawa H. Mammary gland DNA synthesis as 
a limiting factor for mammary tumorigenesis: 
Forum. IRCS Med Sci 5:405-408, 1977. 

4. Nagasawa H. Causes of age-dependency of 
mammary tumour induction by carcinogens in 
rats. Biomedicine 34:9-11, 1981. 

5. Nagasawa H, Yanai R, Taniguchi H. Importance 
of mammary gland DNA synthesis on carcino- 
gen-induced mammary tumorigenesis in rats. 
Cancer Res 36:2223-2226. 1976. 

6. Nagasawa H, Morii S. Prophylaxis of spontane- 
ous mammary tumorigenesis by temporal inhibi- 
tion of prolactin secretion in rats at young ages. 
Cancer Res 41:1935- 1937, 1981. 

7. Nagasawa H, Morii S. Inhibition by early treat- 
ment with CB-154 (bromocriptine) of spontaneous 
mammary tumor development in rats with no 
side-effects. Acta Endocrinol 101:51-55. 1982. 

8. Nagasawa H, Yanai R. Frequency of mammary 
cell division in relation to age: Its significance in 
the induction of mammary tumors by carcinogen 
in rats. J Nat Cancer Inst 52:609-610, 1974. 

9. Russo J, Russo IH. DNA labeling index and 
structure of rat mammary gland as determinants 
of its susceptibility to carcinogenesis. J Nat 
Cancer Inst 61:1451-1459, 1978. 



10. Nagasawa H. The cause of species differences in 
mammary tumorigenesis: Significance of mam- 
mary gland DNA synthesis. Med Hypoth 
5:499-510, 1979. 

11. Yanai R, Nagasawa H. Inhibition of mammary 
tumorigenesis by ergot alkaloids and promotion of 
mammary tumorigenesis by pituitary isografts in 
adreno-ovariectomized mice. J Nat Cancer Inst 
48:715-719, 1972. 

12. Welsch CW, Gribler C. Prophylaxis of spontane- 
ous mammary carcinoma in C3H/HeJ female mice 
by suppression of prolactin. Cancer Res 
33:2939-2946, 1973. 

13. Welsch CW, Gribler C, Clemens JA. 6-methyl-8- 
/3-ergolineacetonitrile (MEA)-induced suppres- 
sion of mammary tumorigenesis in C3H/HeJ 
female mice. Eur J Cancer 10:595-600, 1974. 

14. Mori T, Nagasawa H, Takahashi S. The induction 
of adenomyosis in mice by intrauterine pituitary 
isografts. Life Sci 29:1277-1282, 1981. 

15. Mori T, Nagasawa H, Nakajima H. Strain- 
difference in the induction of adenomyosis by in- 
trauterine pituitary grafting in mice. Lab Anim Sci 
32:40-41, 1982. 

16. Nagasawa H, Yanai R, Taniguchi H, Tokuzen R, 
Nakahara W. Two-way selection of a stock of 
Swiss albino mice for mammary tumorigenesis: 
Establishment of two new strains (SHN and 
SLN). J Nat Cancer Inst 57:425-430. 1976. 

17. Floss HG, Cassady JM, Robbers JE. Influence of 
ergot alkaloids on pituitary prolactin and 
prolactin-dependent process. J Pharmacol Sci 
62:699-715, 1973. 

18. Barflcnecht CF, Rusterholz DB, Parsons JA. Inhi- 
bition of prolactin by ergoline congeners. J Med 
Chem 17:308-312, 1974. 

19. FlUckiger EW. Lactation inhibition by ergot 
drugs. In: Yokoyama A, Mizuno H, Nagasawa H, 
eds. Physiology of Mammary Glands. Tokyo/ 
Baltimore, Japan Sci Soc Press/Univ Park Press, 
PP71-82, 1978. 

20. Snedecor CW. Statistical Methods, 5th ed. Ames, 
Iowa State Univ Press, pp241-253, 1966. 

21. Welsch CW, Nagasawa H. Prolactin and murine 
mammary tumorigenesis: A review. Cancer Res 
37:951-963, 1977. 

22. Welsch CW, Morford LK. Influence of chronic 
treatment with 2-bromo-a-ergocriptine (CB-154) 
on the reproductive and lactational performance 
of the C3H/HeJ female mouse. Experientia 
30:1353-1355, 1974. 

23. Finn CA, Martin L. The cellular response of the 
uterus of the aged mouse to oestrogen and pro- 
gesterone. J Reprod Fertil 20:545-547, 1969. 

24. Mori T. Effects of postpubertal oestrogen iiyec- 
tions on mitotic activity of vaginal and uterine 
epithelial cells in mice treated neonatally with 
oestrogen. J Endocrinol 64:133-140, 1975. 

Received April 21, 1982. P.S.E.BM, V9%l,N^\,\-\\. 



PKOCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 168-173 (1982) 



Blood-Borne Vasoconstrictor Stimulates Release of [^^C]Arachidonate from 
Prelabeled Isolated Perfused Rabbit Gastrocnemius Muscle^ (41493) 

YUGI HAZEYAMA* and RICHARD L. MORETTI' 

Bruce Lyon Memorial Research Laboratory, Children's Hospital Medical Center, 51st and Grove Streets, 

Oakland, California 94609 



Abstract. The effect of a blood-borae vasoconstrictor on vascular resistance and 
anichidonate release was examined using isolated, perfused rabbit gastrocnemius muscle. 
Replacement of blood by perfusion with Tyrode*s solution caused a marked drop in vascular 
resistance. Addition of rabbit plasma or the plasma factor partially purified from human 
blood restored vascular resistance in a concentration dependent manner. The pressor effect 
of the plasma factor was unaffected by concentrations of phentolamine (2.5 /Lig-ml~') and 
sar*-a]a'-angiotensin II (75 /ig* ml~*) sufficient to completely block the effects of norepineph- 
rine and angiotensin II, respectively. However, the pressor effects of the factor were 
inhibited by indomethacin in a concentration-dependent manner. The factor produced a con- 
centration dependent (r = 0.95, n - 12) increase in release of [**C]arachidonate from pre- 
labeled preparations. This release was blocked by indomethacin. These results are consis- 
tent with the hypothesis that the pressor effects of the factor result from its effects upon 
arachidonate metabolism. 



Evidence indicates the existence of an 
unidentified humoral factor which is im- 
portant in the maintenance of basal vascu- 
lar resistance (1-4). Perfusion of the iso- 
lated dog gracilis muscle with a blood-free 
physiological salt solution causes vasodila- 
tion and a concomitant loss of the au- 
toregulatory response to changes in perfu- 
sion pressure. Addition of blood plasma to 
the perfusate restores both the autoregula- 
tory response and vascular resistance. 
Norepinephrine causes vasoconstriction 
but does not restore the autoregulatory re- 
sponse (1). Bohr and Sobieski obtained a 
substance from plasma which produces 
contraction in vascular smooth muscle 
strips isolated from a variety of sites (2). 
We have partially purified a substance from 
blood plasma which restores vascular re- 



' Supported by NIH Grants HL22946, HL19218, 
and RR05467. Dr. Hazeyama was a recipient of an 
Advanced Research Fellowship from the American 
Heart Association, California Affiliate. 

^ Present address: Center for Health Sciences. 
Chandler Laboratories, BIdg. 17, Lehigh University. 
Bethlehem. Pa. 18105. 

•'' To whom reprint requests and correspondence 
should be addressed. 



sistance and the autoregulatory response in 
isolated rabbit hearts perfused with Ty- 
rode's solution (3, 4). 

The mechanism of action of this plasma 
pressor has not been completely delineated. 
Bohr and Johansson found the action of 
their plasma pressor to be unlike that of a 
variety of vasoactive agents (norepineph- 
rine, angiotensin II, histamine, serotonin, 
and vasopressin) (5). We have found that 
the pressor effects of the plasma factor can 
be blocked with inhibitors of prostaglandin 
synthesis such as indomethacin (4). Addi- 
tionally, the partially purified factor stimu- 
lates the conversion of arachidonate to 
prostaglandins and thromboxanes m isolated 
perfused rabbit hearts (3,4), human platelet 
suspensions (6), and cell-free enzyme prep- 
arations (6-8). The factor also stimulates 
the conversion of arachidonate to 12- 
hydroperoxyeicosatetraenoic acid [12- 
HPETE] via the reaction catalyzed by 
lipoxidase (6). The factor alters the relative 
amounts of products formed from arachi- 
donate, increasing formation of vasocon- 
strictors, such as prostaglandin Fao, with 
respect to vasodilators, such as prostaglan- 
din I2. This last effect could result from 
factor-stimulated synthesis of 12-HPETE 



168 
OOJ7-9727/82/l00168-06$0l.00/0 
Copyright re-) 1982 by the Society for Experimental Biology and Medicine. 
Ut rights reserved. 



BLOOD-BORNE FACTOR RELEASES ARACHIDONATE 



169 



which inhibits prostaglandin I2 synthase (9). 
The increase in the ratio of vasoconstrictors 
to vasodilators might account for the pres- 
sor activity of the factor. 

The factor may also increase the release 
of arachidonate from endogenous stores. In 
isolated rabbit hearts the factor increases 
prostaglandin Ej-like activity as assayed 
biologically with rat stomach fundus strips 
(3). Since exogenous arachidonate was not 
used in these experiments, prostaglandins 
must have been formed from endogenous 
arachidonate. The rate determining step in 
prostaglandin synthesis is the release of 
arachidonate from endogenous stores. 
Hence, the factor must have stimulated re- 
lease of arachidonate. 

To confirm this, we examined the effect 
of the factor on release of arachidonate in 
rabbit gastrocnemius muscle prelabeled 
with [^^C]arachidonate. Additionally, we 
compared the pressor effect of the factor 
with other vasoactive agents. 

Methods. Chemicals, Stock solutions of 
all drugs and other vasoactive agents were 
prepared on the day of use. Norepinephrine 
(Levarterenol bitartrate, 0.2%, injection) 
was obtained from Winthrop; indigo car- 
mine (0.8% injection) from Hynscott Wes- 
cott and Dunning; angiotensin II, his- 
tamine, serotonin, vasopressin, arachidonic 
acid and indomethacin from Sigma; phen- 
tolamine mesylate (injection) from Ciba; 
sar^-ala^-angiotensin II (saralasin) from 
Vega Biochemicals and rabbit serum albu- 
min (Cohn fraction V) from ICN Phar- 
maceuticals. ['^C]Arachidonic acid (321 
Ci-g"^) was purchased from New England 
Nuclear. 

Indomethacin (100 mg) was dissolved in 
2.0 ml 95% ethanol (37°) and diluted to 10 
ml with modified Tyrode's solution (137 
mAf NaCl, 4 mAf KCl, 1.8 mAf CaClj, 0.5 
mAf MgS04, 0.35 mAf Na2HP04, 18 mAf 
NaHCOs, 5.6 mAf glucose) containing 35 
mg NajCOa. Radiolabeled arachidonate 
was dissolved in the modified Tyrode's 
solution containing factor-free (4) albumin 

(2.5mgml-'). 

Plasma factor. The plasma factor was 
prepared as described previously (4) except 
that it was obtained from human blood in- 



stead of rabbit blood. The factor is normally 
bound to albumin in blood plasma but can 
be extracted from albumin or plasma with 
chloroform methanol (2: 1) and purified with 
thin-layer chromatographic techniques. 
The extraction procedure yields 60 to 70% 
of the original pressor activity in plasma. 
The partially purified factor was dissolved 
in methanol and stored at -20°. Before use, 
the methanol was evaporated with nitrogen 
and the residue dissolved in modified 
Tyrode's solution containing factor-free al- 
bumin (2.5 mgml"*). For convenience, 
factor was reconstituted to the original vol- 
ume from which it was extracted. Final 
factor concentrations are expressed as a 
percentage of the reconstituted stock solu- 
tion. 

Skeletal muscle and heart preparations. 
New Zealand male rabbits (2.2 to 3.2 kg) 
were anesthetized with sodium pentobar- 
bital (40 mgkg"*). The anesthetic was 
supplemented as required. A tracheotomy 
was performed and the lungs were mechan- 
ically ventilated during surgery. The skin 
of the right leg was removed and the 
gastrocnemius muscle exposed. All blood 
vessels except the major artery and vein 
which supply and drain the muscle were 
ligated. Heparin (1000 unit- kg"*) was ad- 
ministered before cannulation of the ar- 
tery and vein. The excised muscle was 
placed in a humidified chamber and per- 
fused with the modified Tyrode's solution 
(pH 7.4, 3r , saturated with 95% O2, 5% CO2). 

Flow of the perfusate was maintained 
with the aid of a constant speed peristalic 
pump (Cole-Palmer). Pressure was mea- 
sured with a pressure transducer (Statham 
P23Db) and recorded with a pen chart re- 
corder (Cole-Palmer). Muscles were per- 
fused for one hour to remove all residual 
blood before testing. In experiments using 
indomethacin, substances were tested in 
the following sequence: plasma, purified 
factor, indomethacin. Following the exper- 
iments the integrity of the vascular bed 
was examined by injecting indigo carmine 
solution (0.8%). Data from preparations in 
which partial blockage of the vascular bed 
was detected were discarded. Rabbit hearts 
were isolated and perfused as described 



170 



BLOOD-BORNE FACTOR RELEASES ARACHIDONATE 



previously (3, 4). Briefly, the hearts were 
perfused through an aortic cannula with 
modified Tyrode's solution at 37°, pH 7.4, 
saturated with 95% O2, 5% COj. Flow was 
adjusted to attain a basal pressure of 60 mm 
Hg. Aortic pressure was measured and re- 
corded as in the experiments with the gas- 
trocnemius muscle preparations. 

In the experiments comparing the effect 
of the plasma factor to that of known vaso- 
active agents, the agonists were tested by 
bolus injection whereas the antagonists 
were infused at constant rate with the aid of 
a syringe pump (Harvard Instruments). 

Radiolabeled prostaglandins and other 
products synthesized from ["CJarachido- 
nate were extracted from the venous 
efHuent with ethyl acetate and identifled by 
thin-layer chromatography as previously 
described (7). Identification was substan- 
tiated by comparing the migration rates of 
the free acids and their methylated deriva- 
tives to those of authentic standards using 
several solvent systems (7). 

Results. Pressor effects. As perfusion 
with Tyrode's solution was started vascular 
resistance, which had dropped to a low 
value during cannulation of the artery, in- 
creased toward normal in vivo values. 
However, as the vascular system in the 
gastrocnemius muscle was purged of blood, 
vascular resistance decreased markedly, 
demonstrating that the net effect of 
switching from blood perfusion to perfusion 
with Tyrode's solution in this preparation is 
a decrease in vascular resistance (Fig. 1). 
When vascular resistance reached basal 
levels the perfusion pump speed was ad- 
justed to produce a pressure of 30 mm Hg. 



"■0 5 10 15 

Time (minutes) 

Fig. 1. Changes in perfusion pressure associated 
with the beginning of perfusion with Tyrode's solution 
in rabbit gastrocnemius muscle. Recorder tracing from 
one muscle preparation typical of all (15) preparations 
tested. Flow rate, 25 mlmin '• 100 g'. 




1.0 100 tooo 

Concentration (percent) 

Fig. 2. Increases in vascular resistance produced 
by infusion of plasma and purified factor in isolated 
rabbit gastrocnemius muscle. Flow rate, 25 ± 4 
mlmin~*' 100 g~^ initial pressure = 30 mm Hg, resis- 
tance = pressure (mm Hg)ml~''min*100 g, mean ± 
standard error. Data from six gastrocnemius muscle 
preparations. 

The basal flow rate was 24.6 ± 3.9 ml- 100 
g"*min"* (mean ± standard error, 15 gas- 
trocnemius preparations). Infusion of rabbit 
plasma or partially purified human plasma 
factor caused a concentration dependent 
increase in vascular resistance (Fig. 2). 

Comparison of factor with other vasoac- 
tive agents. Among the vasoactive agents 
tested (norepinephrine, angiotensin II his- 
tamine, serotonin, and vasopressin) only 
norepinephrine, angiotensin II, and his- 
tamine produced vasoconstriction in gas- 
trocnemius muscle preparations (Fig. 3). 
Histamine produced vasoconstriction in 
high doses in gastrocnemius muscle but 



A Typico 


R«spon«c& 


A 


kK 


^ I J^- 


^Vv 


Foctof 

(imica 

B Phtfilo 


ADH AOH 
) ftO>uU) (TO«U) 

omine 12 5wgAni) 
NOR 


5HT HtST 
(0^cng) (S^g' 

C Sorolosin 

Foe 'or 

(lT»i»q) 


NOR AS 

(05>ig»(OI>«q) 

_1_ 


Factor 
(ImieQ) 


An 

(OlAifl) 



Fig. 3. Vascular effects of agonists and antagonists 
in isolated rabbit gastrocnemius muscle. Agonists 
were administered by bolus injection. Antagonists 
were infused at constant rates. Abbreviations: ADH, 
vasopressin; 5-HT. serotonin; HIST, histamine; NOR, 
norepinephrine; All, angiotensin II. Recorder tracings 
from a single preparation are illustrative of the re- 
sponses produced in all four muscle preparations 
tested. One milliliter equivalent (ml eq) is the amount 
extracted from one milliliter of blood plasma. 



BLOOD-BORNE FACTOR RELEASES ARACHIDONATE 



171 



produced vasodilation in isolated rabbit 
hearts (Fig. 4). Serotonin also produced 
coronary vasodilation. The effects of nor- 
epinephrine and angiotensin II were com- 
pleted blocked by phentolamine (2.5 
/xgml"*) and saralasin (75 /Ltgml"*), re- 
spectively. Neither of these inhibitors had 
any effect upon factor-induced vasocon- 
striction (Fig. 3). 

Effects on arachidonate metabolism. To 
study the release and metabolism of 
arachidonate the gastrocnemius muscle 
preparations were prelabeled with ["CJara- 
chidonate. Radiolabeled arachidonate (5 
/LiCi in 1.0 ml) was injected into the perfus- 
ate in five aliquots at 2-min intervals. More 
than 90% of the label was retained by the 
preparation. Collection of samples was 
started S min after the last injection. 
Effluent (14 ml) was collected in a 4-min 
period. The radiolabeled substances iden- 
tified in the venous effluent were arachi- 
donate, prostaglandin E2, and the stable de- 
rivative of prostaglandin Ij, 6-keto-prosta- 
glandin Fi^. Because of the low levels of 
prostaglandin E2 and 6-keto-prostaglandin 
Fto recovered from the venous effluents 
using this prelabeling technique, effects of 
the factor on venous levels of these prod- 
ucts could not be determined. Although 
both these products were detected in the 
venous effluent of all four of the muscle 
preparations tested, factor-induced statisti- 
cally significant effects were not obtained. 
However, a correlation (r = 0.95, n = 12) 
between factor-induced vasoconstriction 
and release of [*^C]arachidonate was seen 
(Fig. 5, Table I). The release of arachidon- 
ate was not due to vasoconstriction per se 
because there was no correlation (r = 0.44) 
between the vasoconstriction produced by 



I 



iK- V^ 



Foctor 
(0.4mttq) 



5-HT 
(lOpg) 



HIST HIST 
(S/jq) (IO>ug) 



Fig. 4. Effects of factor, serotonin (5-HT), and 
histamine (HIST) on vascular resistance in isolated 
perfused rabbit hearts. Recorder tracings from one 
heart are typical of those seen in the five isolated 
hearts tested. Flow rate, 20 ml min"'. 



4 








Ui 






* • 










< 








Z 5 








0- ^ 








Q -J 








5S 




* 




< ^ 








0: 5 








<82 




* ,' 




0^ 




* 




^t 








2 ' "^ 




























* norepi 


o: 




* * 

* 


. toe for 
1 



RESISTANCE (% OF INITIAL VALUE) 

Fig. 5. Correlation between release of ["Cjarachi- 
donate and factor-induced vasoconstriction. For fac- 
tor: y = \.\X + 13, r = 0.95 for norepinephrine: y - 
0.6Y + 108, r = 0.44. Data from four isolated gastroc- 
nemius muscle preparations. 



norepinephrine and the release of [*^C]- 
arachidonate (Fig. 5). 

Indomethacin counteracts factor-induced 
vasoconstriction and prostaglandin synthe- 
sis in isolated rabbit hearts (3, 4). In gas- 
trocnemius muscle preparations indometh- 
acin had little effect on vascular resistance 
in the absence of factor, but did suppress 
factor-induced vasoconstriction in a con- 
centration-dependent manner (Table II). 
Inhibition was proportional to the loga- 
rithm of indomethacin concentration {y 
= 36.3Lr -h 6.9, r = 0.97, n = 60). The in- 
domethacin concentration for 50% inhibi- 
tion was 15 ijlM, a period of about 20 min 
was required for stabilization of vascular 
resistance after indomethacin infusion was 
started. After stabilization ['^CJarachidon- 
ate was administered by bolus injection 1 
min before sample collection was started. 
Indomethacin reduced radiolabeled venous 
*^C-prostaglandins both in the presence and 
absence of factor. It is of particular interest 
that indomethacin blocked the factor- 
induced increase in release of ['^CJarachi- 
donate (Table I). 

Discussion. Our results show that the 
factor which restores coronary vasocon- 
striction in isolated rabbit hearts (3, 4) also 
restores vasoconstriction in skeletal mus- 
cle. Our results confirm those of Bohr and 
Johansson (5) which show that the action of 
the plasma factor is unlike that of a variety 
of vasoactive agents. 

The results reported herein substantiate 



172 



BLOOD-BORNE FACTOR RELEASES ARACHIDONATE 



TABLE I. Effect of Vasoactive Factor and Indomethacin on Release of [**C)Arachi donate 
FROM Labeled Rabbit Gastrocnemius Muscle 



Factor 








concentration 


Indomethacin 


Vasoconstriction 


Arachidonate 


(%) 


(niM) 


{% of control) 


(% of control) 








100 


100 


15 





212 ± 28 


174 ±9 


35 





285 ± 18 


326 ±23 


35 


0.2 


132 ±23 


30±5 



Note. Repeated measurements were made on four muscle preparations, means ± standard error. 



results we obtained using isolated rabbit 
hearts (3) which suggest that the factor 
stimulates release of arachidonate from en- 
dogenous stores. In this regard the relation- 
ship between indomethacin and factor is 
important. Indomethacin counteracts the 
effects of the factor on the cyclooxygenase 
reaction in a competitive manner (7). In- 
domethacin has been shown to block the 
release of arachidonate by phospholipases 
(10). Hence, indomethacin and the plasma 
fector have opposing effects on both the 
release of arachidonate and its conversion 
to endoperoxides. This finding is compati- 
ble with recent evidence for enzymatic 
coupling of phospholipase A^ and cyclo- 
oxygenase activities (11). 

The correlation between factor-induced 
arachidonate release and factor-induced 
vasoconstriction is consistent with the hy- 
pothesis that vasoaction of the factor re- 
sults from its effects upon arachidonate 
metabolism. In addition to stimulating the 
release of arachidonate and its conversion 
to endoperoxide prostaglandins via the 
cyclooxygenase reaction (3, 4) the factor 
stimulates conversion of arachidonate to 



12-HPETE by lipoxidase (6). This hydro- 
peroxy fatty acid inhibits the formation of 
the vasodilator prostaglandin Ij from en- 
doperoxide prostaglandins (9). Lipoxidase 
activity has been found in vascular tissue 
(12). Hence the factor stimulates release of 
arachidonate but can simultaneously divert 
it from the formation of prostacyclin, a va- 
sodilator, to vasoconstrictor prostaglandins 
(e.g., PGF,„). 

Administration of exogenous arachidon- 
ate can produce vasodilation, presumably 
through formation of prostaglandin I, (13). 
However, we have shown that the plasma 
factor attenuates arachidonate-induced 
vasodilation while concurrently increasing 
the ratio of prostaglandin F^ to prostaglan- 
din I2 which is formed (8). Arachidonate is 
transformed to prostaglandins, throm- 
boxanes, hydroperoxy fatty acids, and 
leukotrienes (14). Both vasodilators and va- 
soconstrictors can be formed simulta- 
neously, depending upon the activity of the 
enzymes present. The plasma factor has a 
marked effect on the activity of some of 
these enzymes. Thus, it could act by main- 
taining a balance between the vasocon- 



T.ABLE 11, IVDOMETHACIN INHIBITION OF Fm TOR-ISDl CED VASOCONSTRICTION 



Indomethac 


in 




Factor concentration \^r » 




3.5 


-.0 15 


35 



^ 






5: r : 

84 r 14 
*M r 13 


Percentage inhibition 



42 r : 28 r 3 

^^ r i: ^8 r 11 

98 r 14 8*^ r 11 



37 r 4 

65r 24 
83 r 6 



\,:e Percenidge inhibition \fcas calculated bv dividing the factor- induced increase in vascular rcsistuice in 
the presence oi indon>eihacin b> the increase in the absence of indomethacin. subtracting the quotient from 1.0. 
«d muliipKing that difference b> 100 Data from four gastrocnemius preparations tmcan :r standard error). 



BLOOD-BORNE FACTOR RELEASES ARACHIDONATE 



173 



strictors and vasodilators formed from 
arachidonate. 

1. Stainsby WN. Autoregulation in skeletal muscle. 
Circ Res 14(Suppl I): 139- 147, 1964. 

2. Bohr DF, Sobieski, J. A vasoactive factor in 
plasma. Fed Proc 27:13%- 1398. 1968. 

3. Moretti RL, Abraham S, Ecker RR. The stimula- 
tion of cardiac prostaglandin production by blood 
plasma and its relationship to the regulation of 
coronary flow in isolated isovolumic rabbit hearts. 
Circ Res 39:231-238, 1976. 

4. Moretti RL, Abraham S. Stimulation of micro- 
somal prostaglandin synthesis by a blood plasma 
constituent which augments autoregulation and 
maintenance of vascular tone in isolated rabbit 
hearts. Circ Res 42:317-323, 1977. 

5. Bohr DF, Johansson B. Contraction of vascular 
smooth muscle in response to plasma: Compari- 
son with response to known vasoactive agents. 
Circ Res 19:593-601, 1966. 

6. Moretti RL, Lin CY. Coronary vasoconstrictor 
extracted from blood plasma stimulates platelet 
lipoxidase activity. Prostaglandins 19:99-108, 
1980. 

7. Moretti RL, Abraham S. Stimulation of micro- 
sonud prostaglandin synthesis by a vasoactive 
material isolated from blood plasma. Prostaglan- 
dins 15:603-622, 1978. 

8. Moretti RL. Blood borne vasoconstrictor lowers 



ratio of prostaglandin I^ to other products formed 
from arachidonate. Prostaglandins Med 6: 
223-232, 1981. 
9. Moncada S, Gryglewski RJ, Bunting S, Vane JR. 
A lipid peroxide inhibits the enzyme in blood ves- 
sel microsomes that generates from prostaglandin 
endoperoxides the substance (prostaglandin X) 
which prevents platelet aggregation. Prostaglan- 
dins 12:715-737, 1976. 

10. Franson RC, Eisen D, Jesse R, Lanni C. Inhibi- 
tion of highly purified mammalian phospholipases 
As by non-steroidal anti-inflammatory agents. 
Biochem J 186:633-636, 1980. 

11. Erman A, Azuri R, Raz A. Enzymatic coupling of 
phospholipase A, and prostaglandin synthetase 
activities in subcellular-fractions from renal rabbit 
medulla. Biochem J 201:635-640, 1982. 

12. Greenwald JE, Bianchine JR, Wong LK. The pro- 
duction of the arachidonate metabolite HETE in 
vascular tissue. Nature (London) 281:588-589, 
1979. 

13. Dusting GJ, Moncada S, Vane JR. Vascular ac- 
tions of arachidonic acid and its metabolites in 
perfused mesenteric and femoral beds of the dog. 
Eur J Pharmacol 49:65-72, 1978. 

14. Samuelsson B, Borgeat P, Hammarstrom S, Mur- 
phy RC. Introduction of a nomenclature: Leuko- 
trienes. Prostaglandins 17:785-787, 1979. 

Received April 23, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 174-178 (1982) 

Effects of Garlic Products on Lipid Metabolism in Cholesterol-Fed Rats^ (41494) 

MYUNG S. CHP 

Department of Home Economics, Alcorn State University, Lorman, Mississippi 30906 



Abstract. The effect of garlic prepared in several forms on lipid metabolism was studied 
in male rats fed a diet containing \% cholesterol. Garlic was supplemented at 1% of the diet 
as fresh garlic in forms of ethanol extracted garlic residue, ethanol extract of garlic, whole 
garlic, and autoclaved garlic. Diets were fed for 4 weeks from 6 weeks of age. The 
supplementation of garlic products except ethanol-extracted garlic residue reduced plasma 
and liver cholesterol levels. The reduction in the plasma cholesterol by feeding garlic prod- 
ucts was in very low density lipoprotein and low-density lipoprotein cholesterol fractions. 
Animals fed diets supplemented with garlic decreased liver glucose-6-phosphate dehydro- 
genase and malic enzyme activities and also reduced the liver weight, inquinal adipose 
tissue weight, liver total lipids, and plasma triglycerides. The hypocholesterolemic activity 
of garlic was contained in the ethanol extract and stable when autoclaved at 120° for 1 hr. 



The blood cholesterol level has been re- 
ported as an independent risk factor con- 
tributing to the development of coronary 
heart diseases (1-3). Recent studies have 
shown that essential oil from garlic or onion 
reduces the blood cholesterol level in hu- 
mans (4, 5) and animals (6, 7). The essential 
oil of garlic or garlic extract prevented lipid 
accumulation in the aorta and showed pre- 
ventive effects against pathogenic athero- 
sclerosis in rabbits fed an atherogenic diet 
(6, 7). An oral administration of garlic to 
human subjects depressed platelet aggrega- 
tion (5, 8) and the blood glucose concentra- 
tion (9). Garlic is used as a flavoring agent 
world wide. Garlic also has been known to 
have medicinal properties in oriental coun- 
tries. Lyophilized garlic powder added at 
2% level to a diet containing 1% cholesterol 
decreased the cholesterol and lipid levels in 
the blood and liver by increasing fecal 
excretion of neutral and acidic sterols in 
rats (10). 

Garlic can be processed into different 
products and used. The objective of this 
study was to determine the effects of 



' Supported in part by Research Grant 516-15-158 
from the Cooperative State Research Services, 
USDA. 

^ Present address: Human Nutrition Laboratory, 
Lincoln University, Jefferson City, Mo. 65101. 



selected garlic products on lipid metabolism 
in rats including the determination of each 
lipoprotein cholesterol fraction and hepatic 
lipogenic enzyme activities. 

Materials and Methods. Male Sprague- 
Dawley rats (Charles River Breeding Lab- 
oratories, Wilmington, Mass.) were indi- 
vidually housed in stainless-steel cages 
in a room maintained at 22° -24'' with about 
50% relative humidity. The room was light- 
ed from 06:00 to 18:00 hr. The composi- 
tion of a basal diet (control group) con- 
taining 1% cholesterol is given in Table I. 
Garlic bulbs were obtained from a commer- 
cial source, peeled ofT, washed and ground 
by a blender for 2 min with an addition of 
water. A portion of ground garlic was ly- 
ophilized and used as whole garlic. The 
moisture contents of the fresh garlic bulb 
and dried garlic powder were 69.1 and 
1.1%, respectively. Another portion of 
ground garlic was refluxed with 5()% aqueous 
ethanol (v/v) in a 60° water bath for 2 hr, 
filtered, and ethanol was evaporated with N 
gas flow. The resulting solution was con- 
centrated by lyophilization and used as gar- 
lic extract (22.1% of whole fresh garlic). 
The filtered garlic particles (8.9% of whole 
fresh garlic) were lyophilized and u®ed as 
garlic residue. A portion of ground garlic 
was autoclaved for 1 hr at 120°, lyophilized 
and used as autoclaved garlic. 

There were five dietary treatments: (1) 



174 

QOJ7-9727/82/100174-05$0].00/0 

:^yright (a) 1982 by the Society for ExperimentaJ Biology and Medicine. 
*^ nsAfs regerved. 



GARLIC PRODUCTS AND LIPID METABOLISM 



175 



TABLE I. Composition of the Basal Diet 



Ingredient 


g/lOOgdiet 


Casein" 


20.00 


Dextrose" 


25.00 


Com starch" 


40.35 


Com oil* 


5.00 


Cellulose"^ 


2.50 


Mineral mix"-' 


4.00 


Vitamin mix'* 


2.00 


DL-Methionine" 


0.15 


Cholesterol* 


1.00 



" ICN Phamiaceuticals, Inc., Cleveland. Ohio. 

* Tocopherol stripped, ICN Pharmaceuticals, Inc. 
** Brown and Co., Boston, Mass. 

"Supplied by J. T. Baker Chemical Co., Phillips- 
burg, N.J. 

" Provided per kilogram diet; CaCOs, 7.5 g; CaHPG^- 
2H,0, 8.2 g; CuSO^ 5H2O, 0.023 g; Fe(C«H„Or)„ 
0.9 g; KIO„ 0.01 g; K.HPO^, 13.0 g; MgS04 7H,0, 
4.8 g: MnCO,, 0.2 g; NaCl, 4.5 g; Zn CO3, 0.052 g. 

' Vitamins except tocopherol, retinyl palmitate and 
cholecalciferol supplied by General Biochemicals, 
Chagrin Falls, Ohio; tocopherol, retinyl palmitate, 
cholecalciferol supplied by Hoffman- LaRoche, Nut- 
Icy, N.J. 

' Provided per kilogram diet; thiamin* HCl, 10 mg; 
riboflavin, 10 mg; nicotinic acid, 40 mg; calcium panto- 
thenate, 30 mg; folic acid, 3 mg; inositol, 25 mg; biotin, 
0.2 mg; vitamin B-12, 0.02 mg; menadione, 2 mg; 
retinyl palmitate, 10,000 lU; cholecalciferol, 2000 lU; 
DL-a-tocopheryl acetate, 120 lU; choline chloride, 
1500 mg. 

* Eastman Kodak Co., Rochester, N.Y. 



basal diet (control), (2) control plus garlic 
residue, (3) control plus garlic extract, (4) 
control plus whole garlic, (5) control plus 
autoclaved garlic. Garlic products were 
added to the diet equivalent to 2% of a diet 
as fresh garlic (w/w) at the expense of com 
starch. Rats were fed a stock diet for the 
2- week preexperimental period. At 6 weeks 
of age, experimental diets and water were 
provided ad libitum for the next 4 weeks. 
Each dietary treatment was randomly as- 
signed to 10 individually caged rats. The 
body weight and food consumption were 
determined weekly. 

Chemical assay. Blood samples were 
taken into heparinized Vacutainer (Beck- 
ton-Dickinson, Dickinson and Co., Ruther- 
ford, N.J.) tubes by cardiac puncture; the 
rats were in a weakly anesthetized state 
following phenobarbital administration (af- 
ter fasting for 16 hr). Immediately after 



the blood sampling the liver was excised, 
washed in chilled saline solution, blotted 
and cooled in crushed ice. A part of liver 
was used for enzyme assay and the rest of 
the liver was kept at -35°C until analysis 
of lipid components. The inquinal fat pads 
in both sides were removed at around caput 
epididymis and weighed. 

Plasma triglycerides were analyzed en- 
zymatically using Tri-Es (Harleco, Division 
of American Hospital Supply Co., Gibson, 
N.J.). Plasma lipoprotein cholesterol was 
separated by ultracentrifugation (11) and 
plasma total and each lipoprotein choles- 
terol fraction was determined using Beck- 
man Cholesterol Analyzer-2 and enzymatic 
cholesterol assay kit (Beckman Instruments, 
Inc., FuUerton, Calif.). Liver lipids were 
determined gravimetrically after extraction 
according to the method of Folch et al, 
(12) and liver cholesterol was analyzed by 
the method of Kim and Goldberg (13). 

Enzyme assay. Approximately 1.0 g of 
liver was homogenized in 9 ml of 0.15 M 
KCl and centrifuged at 20,000^ for 1 hr. The 
resulting supernatant was used for deter- 
mining enzyme activity and protein. Glucose- 
6-phosphate dehydrogenase (G-6-PDH) 
(EC 1.1.1 .49) was assayed by the method of 
Romberg and Horecker (14) and malic en- 
zyme (ME) (EC 1.1.1.40) was assayed by 
the method of Ochoa (15). The reaction was 
initiated by adding the enzyme source to the 
medium. The assay was conducted at 30** by 
observing changes in absorbancy at 340 nm 
with a Beckman Model-25 Spectrophotom- 
eter (Beckman Instruments, Inc.). The 
protein content of the supernatant frac- 
tion was determined by the method of Low- 
ry et al, (16). Enzyme activity was expressed 
as units per milligram protein where a unit 
was the amount of enzyme which converted 
one nanomole of substrate per minute at 30**. 

Statistical analysis. The data were 
analyzed statistically using the analysis of 
variance technique and the least significant 
difference procedure was used in compari- 
sons of treatment means (17). 

Results. Table II and Fig. 1 summarize 
the results of the experiment. Rats fed the 
diet supplemented with garlic extract con- 
sumed less food than the control and whole 



176 



GARLIC PRODUCTS AND LIPID METABOLISM 



12 



o 
u 

as 
> 



(A 

z 
o 
u 

Q 
O 

< 

O 

H 
Z 
O 



-J 

< 
O 
u. 
O 



U 



CQ 
< 

I- 



t 



la 






^!J 

^i 






c 
o 
U 



09030303030903090090903 



+1 +1 



o — — o — o 

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 






0303030303030303U030303 
O «o O 5 0\ r- rn 

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 



»*» ^ j^ eO P - ; -,_,.,.„ 

o — — o6r4ooo\— o— G> 



030303030303030300030303 

> S — CNJ p »*» S 00 

_ iooo\— o— G> 

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 



0Q 0Q 0Q 03 0Q 0Q 

<<<<<<<<<<<< 

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 
r^ r^ IN t^ r: ^ 9 a '^ S; rs 00 



<<<<<<<<<<<< 

>ooo oooov^^-i^NOfn 
v^"^^^CN^rJ^^Or^p 

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 
w^r^^os — r^ — — ooooQrp 



^6| 



E 

C/3 



-II 
•S-o E 

^ U. CU 



21 
i2 » 

Ss 
*^ "5 i^ 

^ « .2P 

2«i: C 

eo eo ^ 

^ S b 

(/) «« *f 

S cu J 



'I 



> ^ 



*> c *i 

^ 6 - 

«2 "^ *3 

3 1) 1) 

C** > > 



f ^- 
S E 

*> = g 

• o5 



J J J 



o 
V 

e 



e 



s 



CO 

c 

s 



U M 

o *- 









c 


c 


•2 


.0 


CQ 


cd 






> 


'> 


4> 


4> 


•0 


•0 


"2 


T3 


ka 




CO 




•a 




c 




cd 


2 






;/) 


tn 


+1 


+1 


c 


c 


cd 


cd 


1) 


V 


5S 



o 

E 

N 

e 
o 
o 

E 

JC — JC 

"C - "^ 
^•S « 

|:it 

•^ « o 
.E'S S 

Jc g-a 

00-- , 

I ^5 

> c Q 
*- 'I C^ 

4> O X 

> ? I 



GARLIC PRODUCTS AND LIPID METABOLISM 



177 




Fig. 1. Effects of garlic products on plasma very 
low density lipoprotein (VLDL), low-density lipopro- 
tein (LDL), and high-density lipoprotein (HDL) cho- 
lesterol as a percentage of the plasma total cholesterol. 
■■ , Control; |H , garlic residue; SBB • garlic 
extract; |HR « whole garlic; ^M y autoclaved gar- 
lic. Each bar indicates mean ± standard deviation of 
10 rats except the autoclaved garlic group which is 
mean ± standard deviation of 9 rats. 

garlic groups. Body weight gain of rats fed 
garlic products except garlic residue was 
decreased as compared with rats fed the 
control diet (Table II). The plasma concen- 
trations of triglycerides, total cholesterol, 
and free cholesterol were decreased by the 
supplementation of garlic extract, whole 
garlic, or autoclaved garlic. 

The plasma cholesterol was fractionated 
into each lipoprotein cholesterol and the 
percentage of each lipoprotein cholesterol 
to the plasma total cholesterol is given in Fig. 
1 . The proportion of very low density lipo- 
protein (VLDL) cholesterol to the total 
plasma cholesterol was significantly {P < 
0.05) lower in rats fed the garlic extract, 
whole garlic, or autoclaved garlic diets than 
in animals fed the control diet. Just oppo- 
site results were obtained in the proportion 
of high-density lipoprotein (HDL) choles- 
terol to the plasma total cholesterol. 

The supplementation of garlic extract, 
whole garlic, or autoclaved garlic also re- 
duced the liver weight, inquinal fat pad 
weights, liver total lipid, and liver total and 
free cholesterol (Table II). Rats fed the 
diets added with garlic products except 
garlic residue decreased hepatic G-6-PDH 
and ME activities as compared with control 
animals. 



Discussion. Garlic appeared as an effec- 
tive agent to reduce blood cholesterol and 
triglycerides as well as liver lipids and cho- 
lesterol. The reduction in plasma choles- 
terol was in VLDL and LDL cholesterol as 
observed in a previous study (10). The 
VLDL cholesterol values (mean ± SD) for 
control, garlic residue, garlic extract, whole 
garlic, and autoclaved garlic groups were 
34.3 ± 5.0, 31.0 ± 2.9, 18.4 ± 2.4, 20.0 ± 
2.7, and 20. 1 ± 2.6 mg/100 ml, respectively, 
and LDL cholesterol values (mean ± SD) 
for the same treatments were 37.3 ± 3.6, 
35.5 ± 4.2, 26.1 ± 2.2, 27.8 ± 2.9, and 28.5 
± 3.7 mg/100 ml, respectively. Although 
the actual values of HDL cholesterol 
among treatment groups (3 1 .2 ± 4. 1 , 3 1 .2 ± 
3.0, 30.1 ± 2.8, 30.0 ± 4.2, and 30.6 ± 2.8 
mg/100 ml for control, garlic residue, garlic 
extract, whole garlic, and autoclaved garlic, 
respectively) were similar, the percentage 
of HDL cholesterol to the plasma total 
cholesterol was higher in animals fed the 
garlic products except garlic residue than 
the control group. This was because the 
plasma total cholesterol level was much 
lower in those garlic groups than in the 
control group. The importance of plasma 
cholesterol and lipoprotein concentrations 
in the pathogenesis of atherosclerosis has 
been noted by numerous investigators. In- 
creased levels of total plasma cholesterol 
with increases in LDL or VLDL were as- 
sociated with a greater risk of developing 
coronary heart diseases (1, 2) while high 
concentration of HDL appeared to be pro- 
tective (18, 19). The changes in the propor- 
tions of lipoprotein cholesterol fractions 
may be one reason of protective effect of 
garlic against the atherosclerotic process. 

G-6-PDH and ME activities have been 
shown to be correlated to lipogenic capac- 
ity in various tissues including rat liver (20) 
by supplying substrates for fatty acid syn- 
thesis (21). Garlic appeared to reduce fatty 
acid synthesis by decreasing key enzyme 
activities in supplying substrates and con- 
sequently reduced the lipid levels in the 
liver and plasma. The decrease in food in- 
take in some of garlic-fed rats did not cause 
these metabolic changes since pair-fed ani- 
mals with and without garlic in a previous 



178 



GARLIC PRODUCTS AND LIPID METABOLISM 



Study (10) showed similar results as in the 
present study. The lesser weight gain in rats 
fed garlic products might be the result in 
part of decreases in food intake and in part 
of the metabolic changes due to garlic 
feeding. The observed metabolic changes in 
garlic-fed animals were not likely due to 
hepatic damage: the liver and gastrointesti- 
nal tract of rats were grossly examined for 
any tissue damage and found no visible tis- 
sue damage in controls and garlic groups. 

Garlic was similarly effective in lowering 
cholesterol and lipid levels in the plasma as 
well as in the liver in a form of lyophilized 
whole garlic or ethanol extract of garlic 
suggesting that the active agent in garlic for 
reducing blood cholesterol was contained in 
the garlic extract. The nonsignificance of 
garlic residue, the portion after ethanol ex- 
tract of whole garlic, for hypocholes- 
terolemic activity also confirms the effec- 
tiveness of ethanol extract of garlic. Au- 
toclaving the garlic at 120** for 1 hr did not 
diminish the hypocholesterolemic property 
of garlic. 



1. Fredrickson OS, Levy RL Lees RS. Fat transport 
and lipoproteins — An intergrated approach to 
mechanisms and disorders. N Engl J Med 
276:32-35, 1%7. 

2. Stamler J. Epidemiology of coronary heart dis- 
ease. Med Clin North Amer 57:5-46, 1973. 

3. Gofman JW, Young W, Tandy R. Ischemic heart 
disease, atherosclerosis and longevity. Circula- 
tion 34:679-685, 1966. 

4. Bordia A, Bansal HC. Arora SK, Singh SV. Effect 
of the essential oils of garlic and onion on alimen- 
tary hyperlipemia. Atherosclerosis 21:15-20, 
1975. 

5. Jain RC. Effect of garlic on serum lipids, coagula- 
bility and fibrinolytic activity of blood. Amer J 
Clin Nutr 30:1380- 1381. 1977. 

6. Kritchevsky D. Effect of garlic oil on experimen- 
tal atherosclerosis in rabbits. Artery 1:319-323, 
1975. 

7. Bordia A, Verma SK, Vyas AK, Bhu N. Effect of 
essential oil of onion and garlic on experimental 



atherosclerosis in rabbits. Atherosclerosis 
26:379-386, 1977. 

8. Bordia A, Joshi HK, Sanadhya YK, Bhu N. Ef- 
fect of essential oil of garlic on serum fibrinolytic 
activity in patients with coronary artery disease. 
Atherosclerosis 28:155- 159, 1977. 

9. Jain RC, Vyas CR, Mahatma OP. Hypoglycemic 
action of onion and garlic. Lancet 2:1491, 1973. 

10. Chi MS, Koh ET, Stewart TJ. Effect of garlic on 
lipid metabolism in rats fed cholesterol or lard. J 
Nutr 112:241-248, 1982. 

11. Bronzert TJ, Brewer HB. New micromethod for 
measuring cholesterol in plasma lipoprotein frac- 
tions. Clin Chem 23:2089-20%, 1977. 

12. Folch J, Lees M, Sloane-Stanley GH. A simple 
method for the isolation and purification of total 
lipids from animal tissues. J Biol Chem 
226:497-509, 1957. 

13. Kim E, Goldberg M. Serum cholesterol assay 
using a stable Liebermann-Burchard reagent. 
Clin Chem 15:1171-1179, 1%9. 

14. Komberg A, Horecker BL. Glucose-6-phosphate 
dehydrogenase. In: Colowich SP, Kaplan NO, 
eds. Methods in Enzymology. New York. 
Academic Press, Vol l:p323, 1955. 

15. Ochoa S. Malic enzyme. In: Colowich SP, Kaplan 
NO, eds. Methods in Enzymology. New York, 
Academic Press, Vol I:p739, 1955. 

16. Lowry OH, RosebroughNJ.FarrAL, Randall RJ. 
Protein measurement with the Folin phenol re- 
agent. J Biol Chem 193:265-272, 1951. 

17. Steel RGD, Torrie JH. Principles and Procedures 
of Statistics. New York, McGraw-Hill, p481, 
1960. 

18. Miller GJ. Miller NE. Plasma high density lipo- 
protein concentration and development of isch- 
emic heart disease. Lancet 1:16-19, 1975. 

19. Gordon T, Castelli WP, Hjortland MC, Kannel 
WB, Dawber, TR. High density lipoprotein as a 
protective factor against coronary heart disease. 
Amer J Med 62:707-714, 1977. 

20. Tepperman HM, Tepperman J. Patterns of dietary 
and hormonal induction of certain NADP-Unked 
liver enzymes. Amer J Physiol 206:357-362, 
1964. 

21. O'Hea EK, Leveille GA. Lipogenesis in isolated 
adipose tissue of the domestic chick. Comp 
Biochem Physiol 26:111-117, 1968. 

Received January 22, 1982. P.S.E.B.M. 1982, Vol. 
171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 179-183 (1982) 

Survival of Porcine Embryos after Asynchronous Transfer^"^ (41495) 
W. F. POPE,* t R. R. MAURER,* ^ and F. STORMSHAKt 

*Roman L. Hruska U.S. Meat Animal Research Center, ARS, U.S. Department of Agriculture, 
Clay Center, Nebraska 68933, and ^Department of Animal Science, Oregon State University, 

Corvallis, Oregon 97331 



Abstract. Forty gilts and sows were used to transfer Day 5 embryos into one uterine horn, 
while Day 7 embryos were transferred into the other horn, of Day 6 nonpregnant recipients 
(Day = first day of estnis). The survival of the transferred embryos was determined on 
Day 11 (Expt 1) and Days 60 to 70 (Expt 2). The percentage of Day 5 and 7 embryos 
surviving the transfer procedures on Day 11 was not different, 42 ± 10 and 43 ± 12, 
respectively. However, by midgestation (Day 60) more {P < 0.001) fetuses that developed 
from Day 7 embryos survived than fetuses that developed from Day 5 embryos, 63 ± 8 and 8 
± 7%, respectively. These experiments indicated that the presence of embryos more ad- 
vanced in development caused the demise of younger embryos sometime between Days 1 1 
and 60 of gestation. 



Considerable variation exists in em- 
bryonic development within several poly- 
tocous species. This is not surprising be- 
cause in pigs ovulation extends for a 6-hr 
period (1, 2). Accordingly, the first cleav- 
age division of the fertilized ova occurs 
between 60 and 108 hr after the onset of 
estrus in sows (3). Anderson (4) noted 
marked variation in the morphology of por- 
cine embryos between Days 11 and 13 of 
gestation. Within a uterine horn these em- 
bryos ranged in development from spheri- 
cal and tubular to filamentous. Embryo 
survival was not altered when embryos 
were transferred 1 day from synchrony with 
the recipient (5). However, the biological 
significance of this normally occurring vari- 
ation in embryonic development remains 
unclear. 

Pigs naturally lose 40% of their embryos 
during gestation. Several physiological 



» Technical Paper No. 6157, Oregon, Agr. Exp. Sta. 

* Mention of a trade name, proprietary product or 
specific equipment does not constitute a guarantee or 
warranty by the U.S. Department of Agriculture and 
does not imply its approval to the exclusion of other 
products that may be suitable. 

' The authors are grateful to Dr. Ron Lindvall, Jean 
Gray. Edward McReynolds, and Scott Sholtz for their 
technical assistance. 

* To whom all correspondence should be addressed. 



events, important for survival of the por- 
cine embryo, are closely associated with 
morphological changes of the embryo. 
These events include estrogen synthesis 
(6), luteal maintenance (7), and transuterine 
migration of embryos (8). The possibility 
exists that embryos advanced in their de- 
velopment relative to their litter mates have 
a survival advantage. 

The present experiment was conducted 
to determine if embryos more advanced in 
development migrated further within the 
uterine horns and had a greater chance for 
survival during pregnancy. 

Materials and Methods. Expt 1, Sixteen 
cross-bred gilts and sows, checked daily for 
estrous behavior, were utilized in this ex- 
periment. Recipients, 6 days after the onset 
of estrus, received four to six embryos each 
from Day 5 and 7 donors (Day = first day 
of estrus). Such a procedure allowed estab- 
lishment of pregnancy with embryos 2 days 
apart in age but only 1 day from synchrony 
with the recipient. The uterine horn to 
which the embryos, within an age, were in- 
troduced was randomized. Gilts assigned 
on the appropriate days as donors were 
mated 4 and 24 hr after the onset of estrus. 
To maximize utilization of females, two re- 
cipients were used for each Day 5 and 7 
donor. On Day 11,5 days post-transfer, the 
recipients were slaughtered and the em- 



179 
0037.9721/%2i\Wi\l^Mi5%^\ .^^ 

Copyright © IWlb^ xVft SocxtVj lot 'E.xvwvsnftx^A^ 
AU rigjKu reserved. 



180 



SURVIVAL OF ASYNCHRONOUS EMBRYOS 



bryos recovered by flushing segments (10 to 
20 cm) of the excised uterus with physio- 
logical saline. Age (Day 10 or 12) and loca- 
tion of the embryos and the length of 
uterine segments were recorded. The age of 
the recovered embryos was determined by 
morphology (spherical vs tubular or oblong) 
and size. Recovered embryos classified as 
originating from Day 5 donors, for example, 
were all spherically shaped, measuring 1 to 
2 mm in diameter. Those identified as ori- 
ginating from Day 7 donors were spherical 
to tubular in shape, ranging from 4 to 15 mm 
in length. Only those paired recipients with 
recoverable embryos from each donor, as 
determined on Day 11 of gestation, were 
included in the statistical analysis. 

The distance the embryos migrated (cm) 
was determined as follows: distance = 
{(sum of the length of uterine segments 
traversed by each embryo) - (the total 
number of Day 5 or 7 embryos transferred) 
X [the length of the anterior uterine seg- 
ment ( 10 to 20 cm) containing the respective 
embryos]}. It was necessary to subtract the 
length of the anterior segments because the 
embryos were located in these segments 
before their migration. Distance the Day 5 
or 7 embryos migrated was compared by 
using a least-squares analysis of variance. 

Expt. 2. Twenty-four [eight purebred 
colored (Duroc, Spotted or Hampshire), 
eight purebred white (Yorkshire, Landrace 
or Large White) and eight crossbred] gilts 
and sows were utilized in this experiment. 
Day 6 recipients received embryos in ac- 
cordance with procedures described in 
Expt 1 except donors were mated to col- 
ored or white boars. One recipient was used 
for each Day 5 and 7 donor. Attempts were 
made to balance the number of Day 5 and 7 
embryos transferred into each recipient 
(10.9 ± 0.9 Day 5 and 10.4 ± 1.0 Day 7 
embryos, jc ± SE). Breed of donor (colored 
vs white) was randomized such that four 
Day 5 and four Day 7 colored donors were 
utilized. Fetuses were recovered between 
Days 60 and 70 of gestation, identified by 
skin pigmentation, weighed, and the dis- 
tance between adjacent fetuses was noted. 
One gilt aborted on Day 60, in which case, 
all fetuses were recovered immediately and 



skin pigmentation was noted. Fetal weight 
was subsequently correlated with the dis- 
tance between adjacent fetuses. The per- 
centage of Day 5 and 7 embryos surviving 
to Day 11 (Expt 1) and 60 of gestation was 
compared by a nonparametric Mann- 
Whitney U test. 

To confirm that Day 5 embryos can sur- 
vive in Day 6 recipients in the absence of 
Day 7 embryos, 10 additional recipients re- 
ceived only Day 5 embryos (8.8 ± 0.7 em- 
bryos per uterine horn). Comparisons of the 
percentage of Day 5 and 7 embryos surviv- 
ing to Day 60 (Expt 2) to the percentage of 
Day 5 embryos surviving to Day 60 alone 
were conducted by use of a nonparametric 
Mann- Whitney U test and unpaired / test, 
respectively. 

Embryo manipulation. Embryos were 
collected surgically from donors by flushing 
the anterior half of each uterine horn to- 
ward a catheter (medical grade Teflon tub- 
ing, 1.50-mm i.d., 2.11-mm o.d.) located 1 
to 2 cm posterior to the uterotubal junction. 
The flushing medium (Table I) was similar 
to that utilized by Davis and Day (9) except 
lactate and pyruvate were deleted; antibi- 
otic-antimycotic was substituted for 
penicillin G and streptomycin, and sodium 
chloride increased to maintain physiologi- 
cal osmolarity. Recovered embryos were 
incubated (39°, 95% air- 5% CO2) for not 
more than 30 min before being transferred. 
This transfer procedure consisted of 
effluxing the embryos and medium (100 to 
200 pX) into the uterine lumen (2 to 3 cm 
posterior to the uterotubal junction) 
through a catheter (medical grade Teflon 
tubing, 0.69-mm i.d., 0.99-mm o.d.) tem- 
porarily inserted through the posterior 3 to 
4 cm of the oviduct. Only morphologically 
normal embryos were utilized. Transferable 
Day 5 embryos ranged from late compacted 
morula to blastocysts with a small to 
medium blastocoel and a recognizable inner 
cell mass. The Day 7 embryos had an ex- 
panded blastocoel and a comparably flat- 
tened inner cell mass and thinner zona pel- 
lucida. 

Results and Discussion. Results of Expt 
1 (Table II) indicated no difference in the 
ability of Day 5 and 7 embryos to survive 



SURVIVAL OF ASYNCHRONOUS EMBRYOS 



181 



TABLE L Modified Krebs- 
RiNGER Bicarbonate" 



Ingredient 


g/500ml 


mM 


Naa 


3.500 


119.78 


KCl 


0.178 


4.78 


CaCl, 2H,0 


0.125 


1.71 


KH,P04 


0.081 


1.19 


MgSO« 7H/) 


0.147 


1.19 


NaHCO, 


1.053 


25.00 


Glucose 


0.500 


5.56 


Bovine serum albumin* 


2.000 




Antibiotic - antimycotic <^ 


50,000 units/5 ml 





' Davis and Day (1978). 

* Pentex bovine albumin crystallized. Miles Laboratories. 

<* Antibiotic -antimycotic, lyophilized, Gibco Laboratories. 



for 5 days after transfer. However, by 
midgestation (Day 60) more fetuses that de- 
veloped from Day 7 embryos (Expt 2; Table 
II) survived than fetuses that developed 
from Day 5 embryos (P < 0.001). At Day 60 
only two recipients had fetuses that devel- 
oped from transferred Day 5 embryos. Only 
females that remained pregnant to Day 60 
were included in the data of Expt 2. This 
increased the proportion of Day 7 embryos 
recovered at Day 60 as compared with Day 
1 1 because the percentage of all transferred 
Day 5 vs Day 7 embryos surviving to Days 
1 1 and 60 was 34 vs 44 and 4 vs 36, respec- 
tively. 

More (P < 0.01) Day 5 embryos survived 
to Day 60 in the absence than in the pres- 
ence (42 vs 8%) of Day 7 embryos (Table 
II). Webel et al, (5) observed an equivalent 
survival rate of embryos transferred one 
day from synchrony (49 to 53%). Alterna- 
tively, the percentage of Day 7 embryos 
surviving to Day 60 in the presence of Day 5 
embryos was not different from the per- 
centage of Day 5 embryos surviving to Day 
60 alone (42 vs 63%, respectively). These 
observations confirm that each population 
of transferred embryos can survive alone 
but when forced to cohabit in the uterus, 
fewer younger embryos survived to Day 60. 

Identification of embryos by size in Expt 
1 was more subjective than skin pigmenta- 
tion of the fetuses in Expt 2. Wright and 
Granmier (10) observed a 25-fold increase 
in protein content of the porcine embryos 
between Days 8 and 9 of gestation. This 



QQ 

< 



& 



<§ 



.> c 

3 a 



•a 



• %^ 

Z -S ^ 
S g 
v 2 



> c 

CO O 



II 



•a 
Z -2 ^S 



c g 

2 Cl. CO 

'^'5 2 

2> a 



z.s-s 



+1 +1 



+1 +1 -n 
CM 00 r^ 



o 

E 



O r^ O 

S § £ 
« *^ oJ 

>» ^ >» 

^ 00 /i*. 
^ ^ OS 
^* 00 

® 2 o 
S « rt 
= o^S 

"O S e 

S-S O 
•o 3 t! 

|2| 
S.SS 

^■£18 

^ E.S? S 

Si .6 e J> 

"■ -o "^ Ci. 
u H u D. 

,S ■- C 3 
*** ^ — «« 

^ tJ ^ S 

U M U ^ 

4^ ^ U (C 






It 5 I 
o g a c 

V C U (0 



182 



SURVIVAL OF ASYNCHRONOUS EMBRYOS 



exponential growth of embryos continued 
between Days 9 and 16. Although consider- 
able variation existed, spherical shaped 
Day 12 embryos were larger and contained 
fourfold more protein than spherical Day 10 
embryos (4). Little difficulty was experi- 
enced in the present study in differentiating 
between the transferred Day 5 and 7 em- 
bryos on Day 1 1 . 

Transuterine migration of porcine em- 
bryos occurs between Days 7 and 12 (11). 
Recovered embryos 5 days post-transfer 
were mixed within both uterine horns as 
was previously observed with synchronous 
transfer (12). The older embryos (Day 12) 
failed to migrate further than the younger 
embryos (Day 10) when examined on Day 
11, 160.1 ± 29.6 vs 113.5 ± 16.1 cm, re- 
spectively. However, the distance the em- 
bryos migrated may have been different if 
examined at an earlier time. Because only 
about 40% of the transferred embryos were 
viable on Day 11, differentiating the healthy 
from the dying embryos may have been dif- 
ficult before this time. 

Recovered fetuses ranged in weight from 
60.1 to 274.5 g and crown-rump length 
from 7.5 to 19 cm. However, neither fetal 
weight nor length was associated with the 
age of the transferred embryo (172.7 ± 14.2 
vs 157.0 ± 9.3 g and 14.5 ± 0.9 vs 13.9 ± 
0.4 cm. Day 5 vs Day 7 embryos, respec- 
tively). After the variation in fetal weight 
between recipients was reduced by stan- 
dardizing the weight of the heaviest fetus 
within each litter and then adjusting the 
weight of each remaining litter mates pro- 
portionately, the distance between adjacent 
fetuses was highly correlated with fetal 
weight (r = 0.47, P < 0.01). Rathnasaba- 
pathy et al. (13) observed a similar relation- 
ship with fetuses examined on Day 55 of 
gestation. Knight et al. (14) demonstrated a 
significant correlation of fetal weight to 
placental length suggesting a relationship 
between migration of the porcine embryo, 
outgrowth of the placenta, and the subse- 
quent development of the fetus. 

Although the majority of embryonic loss 
occurs by Day 30 (15-19) it is unknown 
when or why the young embryos (Day 5) 
died between Days 11 and 60 of gestation. 



Because the porcine embryo can elongate 
rapidly, 3 cm/hr (20), the possibility exists 
that older embryos (Day 7) elongated 
sooner and occupied more of the uterus 
than the younger (Day 5) embryos. Ander- 
son (4) observed the inability of embryos to 
overlap each other regardless of the uterine 
space available. Knight et al, (14) observed 
an increase in mortality of crowded fetuses 
between Days 40 and 100 and suggested this 
was due to placental insufficiency. Perhaps 
in this experiment the younger embryos 
(Day 5) died because of placental insuffi- 
ciency as a result of crowding by the older 
embryos (Day 7). 

Another explanation for the loss of the 
younger embryos (transferred at Day 5) 
might include physiological advancement in 
the biochemical development of the recipi- 
ent's uterus such that the younger embryos 
could no longer continue to develop. Beier 
et al, (21) and Adams (22) demonstrated the 
fragile relationship between synchronizing 
the pattern of uterine secretions and the age 
of successfully transferred rabbit embryos. 
Exogenous estrogen extends the length of 
the estrous cycle of nonpregnant pigs (23) 
possibly by altering secretion of uterine 
proteins (24), intrauterine sequestering of 
prostaglandin (25), and/or uterine blood 
flow (26). It is possible the older embryos 
(Day 7), by synthesizing estrogen earlier, 
advanced the secretory pattern(s) of the 
uterus resulting in the demise of the 
younger embryos (Day 5). 

The precise mechanism by which some 
embryos survive and others die in polyto- 
cous species is unclear. These experiments 
indicated porcine embryos more em- 
bryologically advanced have a greater 
chance to survive and may have caused the 
demise of those less embryologically devel- 
oped. 



1 . Burger JF. Sex physiology of pigs. Onderstepoort 
J Vet Res 25(Suppl 2): 1-218, 1952. 

2. Lewis LL. The vitality of reproductive cells. Okla 
Agric Exp Stn Bull 96, 1911. 

3. Oxenreidcr SL, Day BN. Transport and cleavage 
of ova in swine. J Anim Sci 24:413-417, 1965. 

4. Anderson LL. Growth, protein content and dis- 



SURVIVAL OF ASYNCHRONOUS EMBRYOS 



183 



tribution of early pig embryos. Anat Rec 
190:143-153, 1978. 

5. Webcl SK, Peters JB, Anderson LL. Synchro- 
nous and asynchronous transfer of embryos in the 
pig. J Anim Sci 30:565-568, 1970. 

6. Ford SP, Christcnson RK, Ford JJ. Uterine arte- 
rial blood flow and uterine luminal content and 
secretion of oestrone and oestradiol-17/3 on Days 
11, 13, and 15 of the oestrous cycle and gestation 
of sows. J Reprod Fertil 64:185-190, 1981. 

7. Frank M, Bazer FW. Thatcher WW, WUcox CJ. A 
study of prostaglandin F^a as the luteolysin in 
swine. III. Effects of estradiol valerate on pros- 
taglandin F, progestins, estrone and estradiol 
concentrations in the utero-ovarian vein of non- 
pregnant gilts. Prostaglandins 14:1183-1196, 
1977. 

8. Patten BM. Embryology of the Pig, 3rd ed. 
Blakiston, Philadelphia, 1948. 

9. Davis DL, Day BN. Cleavage and blastocyst for- 
mation by pig eggs in vitro. J Anim Sci 
46:1043-1053, 1978. 

10. Wright RW, Jr, Grammer JC. Size variation and 
total protein in porcine embryos collected from 
individual pigs. Theriogenology 13:111, 1980. 

11. Dhindsa DS, Dziuk PJ, Norton HW. Time of 
transuterine migration and distribution of em- 
bryos in the pig. Anat Rec 159:325-330, 1967. 

12. Dziuk PJ, Polge C, Rowson LE. Intra-uterine mi- 
gration and mixing of embryos in swine following 
egg transfer. J Anim Sci 23:37-42, 1964. 

13. Rathnasabapathy V, Lasley JF, Mayer DT. Gene- 
tic and environmental factors affecting litter size 
in swine. MO Agric Exp Stn Bull 615, 1956. 

14. Knight JW. Bazer FW, Thatcher WW, Franke 
DE, Wallace HD. Conceptus development in in- 
tact and unilaterally hysterectomized-ovariecto- 
mized gilts: Interrelations among hormonal status, 
placenta] development, fetal fluids and fetal growth. 
J Anim Sci 44:620-637, 1977. 

15. Longenecker DE, Day BN. Fertility level of sows 
superovulated at post weaning estrus. J Anim Sci 
27:709-711, 1968. 



16. Pope CE, Vincent CK, Thrasher DM. Effect of 
I.C.I. 33,828 and PMS on reproduction in gilts. J 
Anim Sci 27:303A, 1968. 

17. Pope CE, Christcnson RK, Zimmerman-Pope VA, 
Day BN. Effect of number of embryos on em- 
bryonic survival in recipient gilts. J Anim Sci 
35:805-808, 1972. 

18. Bazer FW, Robison OW, Clawson AJ, Ulbcrg LC. 
Uterine capacity at two stages of gestation in gilts 
following embryo superinduction. J Anim Sci 
29:30-34, 1%9. 

19. Webcl SK, Dziuk PJ. Effect of stage of gestation 
and uterine space on prenatal survival in the pig. J 
Anim Sci 38:960-%3, 1974. 

20. Geisert RD, Bazer FW, Brookbank JW. Morpho- 
logical changes association with porcine blasto- 
cyst elongation. J Anim Sci 53(Suppl 1):320A. 
1981. 

21. Bcier HM, Mootz U, Kuhnel, W. Asynchronous 
egg transfer during delayed secretion in the rabbit. 
VII. Inst. Congress for Animal Reproduction, 
Munich, Vol 3:ppl891-18%, 1972. 

22. Adams CE. Asynchronous egg transfer in the rab- 
bit. J Reprod Fertil 35:613-614, 1973. 

23. Gardner ML, First NL, Casida LE. Effect of ex- 
ogenous estrogen on corpus luteum maintenance 
in gilts. J Anim Sci 22:132-138, 1%3. 

24. Geisert RD, Bazer FW, Basha SMM, Roberts 
RM. Quantitative and qualitative aspects of 
uterine protein secretions from pseudopregnant 
and unilaterally pregnant gilts. J Anim Sci 
49(Suppl 1):299A, 1979. 

25. Frank M, Bazer FW, Thatcher WW, Wilcox CJ. A 
study of prostaglandin F^a as the luteolysin in 
swine. IV. An explanation for the luteotrophic 
effect of estradiol. Prostaglandins 15:151-159, 
1978. 

26. Ford SP, Magness RR. Effect of intra-uterine in- 
fusion of estradiol- 17/3 (Eja) on luteal function in 
nonpregnant sows. J Anim Sci 51(Suppl 1):279, 
1980. 

Received May 3, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 184-190 (1962) 

Evidence for an Intestinal Factor Stimulating Hepatic Cholesterogenesis (41496) 
JOSE E. DOS SANTOS,* KANG-JEY HO,t and CARLOS L. KRUMDIECKt' 

*I>epartment of Internal Medicine, Faculdade de Medicina de Ribeirdo Preto, USP, BrazU, Departments of 
f Pathology, and XNutrition Sciences, University of Alabama in Birmingham, and Veterans Administration 

Hospital, Birmingham, Alabama 35294 

Abstract. External diversion of bile leads to an increase in hepatic cholesterol synthesis. 
To study the role of the small intestine in this phenomenon we conducted a series of 
experiments in bile-diverted rats with and without surgical removal of most of the small 
intestine, its upper half or its lower half. The pancreas was preserved intact in aU experi> 
ments. Hepatic cholesterol synthesis at the time of surgery and 24 hr later was measured in 
liver homogenates using [2-*^C]acetate as substrate. Hepatic cholesterogenesis increased 
almost 4-fold 24 hr after biliary diversion in rats with intact intestine, and decreased to 64% 
of the baseline rate in bOe-diverted rats with the jejunum- ileum removed, and to 58% in 
sham-operated animals. To investigate the possibility that the stimulation seen in the bile- 
diverted rats with intact intestine was due to a substance absorbed from the diet, the 
experiments were repeated with animals fed only 10% glucose in water 24 hr prior to 
surgery. Again the rats with intact small intestine showed an increase in hepatic choles- 
terogenesis (6.7-fold) while those with the jejunum-ileum removed and the sham operated 
showed a decrease to 60 and 70% of the baseline rate, respectively. These results support the 
hypothesis that the small intestine produces a factors) that stimulates hepatic cholesterol 
synthesis in response to a drop in the intraluminal content of bile. To narrow the site of 
production of the factors), the effect on hepatic cholesterogenesis of removal of the upper 
or lower half of the smaU intestine in bile diverted animals was studied by comparison to a 
group of bile-diverted intestine-intact controls. As before, the latter showed a 6-fold in- 
crease in hepatic cholesterogenesis. With the upper or lower half of the small intestine 
removed a similar degree (4- to 5-fold) of stimulation was still observed. These results 
suggest that the factors) is produced along the entire length of the jejunum-ileum. 



On the basis of a substantial body of tulated that the intestine produces an inhib- 
suggestive evidence Knimdieck and Ho (1) itory factor of hepatic cholesterogenesis, 
have postulated that the intestine plays a the amount of which is decreased when the 
role in the regulation of hepatic choles- concentration of bile in the lumen de- 
terogenesis by producing a stimulatory creases. If this hypothesis is true the fiinc- 
factor whenever the intraluminal concen- tional integrity of both the intestine and the 
tration of bile acids drops below require- portal circulation would be essential for the 
ments. The factor is presumably trans- elevation of hepatic cholesterogenesis in 
ported to the liver by the portal circulation, bile-diverted animals. Consequently, re- 
According to this hypothesis, the well- moval of the small intestine from a bile- 
documented increase in the rate of hepatic diverted animal ought to completely elimi- 
cholesterol synthesis seen in animals in nate the normally observed elevation of the 
which the flow of bile has been diverted rate of cholesterol synthesis in the liver. In 
away from the intestinal lumen by means of this article we report the effects on hepatic 
a biliary fistula (2) would be the result of an cholesterogenesis of the simultaneous pro- 
increased production of the intestinal stim- duction of a bile fistula and the removal of 
ulatory factor. Conversely, it may be pos- nearly all the small intestine (from the liga- 
ment of Treitz to the cecum), its upper half, 
or its lower half, in rats. The results ob- 



• To whom reprint requests and correspondence tained support the existence of the pro- 

-»»ouid be addressed. posed intestinal Stimulatory factor. 

184 

'^27/g2/100I84'a7$OLOO/0 
-*Ar ^ /9f2 by the Society for txpenmenuJ Bioiogy and Medicine. 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



18S 



Materials and Methods. Animals and 
surgical procedures. Male Sprague- Daw- 
ley rats weighing 200 to 250 g were used in 
all the experiments. The animals were 
maintained in an air-conditioned room with 
alternating 12-hr light and dark cycles; the 
lights were on from 6:00 am until 6:00 pm. 
To minimize confounding effects attribut- 
able to the circadian rhythm of hepatic 
cholesterol synthesis all animals were oper- 
ated and sacrificed between 9:00 and 10:00 
am. The animals were maintained for at 
least one week prior to surgery on a mod- 
ified MIT-200 diet (3) containing by weight 
20% casein, 65% sucrose, 5% com oil, 2% 
agar, and 4% cellulose plus 1% vitamin mix 
and 3% salt mixture. Under ether anes- 
thesia groups of animals were subjected to 
one of the following surgical procedures: 

(a) Liver sampling and biliary diversion: 
After ligating the pedicle, the lateral portion 
of the medial lobe of the liver was excised 
and immediately homogenized as described 
below to determine the baseline rate of he- 
patic cholesterol synthesis. After removal 
of the liver sample the common bile duct 
was cannulated using a polyethylene tube 
(PE-10; o.d. 0.6 mm) which was exter- 
nalized through a small orifice in the ab- 
dominal wall and securely tied to the skin. 

(b) Liver sampling, biliary diversion, 
and removal of the small intestine: The 
animals in this group underwent the same 
procedures to remove a sample of liver and 
to establish a biliary fistula as those in 
group (a). In addition, most of the small in- 
testine was removed along its mesenteric 
attachment leaving in place the duodenum 
and approximately 1 cm of the distal ileum. 
The duodenum and terminal ileum were 
connected with a short plastic tube (o.d. 4 
mm) to prevent the development of gastric 
distention. This operation leaves the pan- 
creas intact and does not alter the venous 
drainage of the pancreatic circulation into 
the portal system. 

(c, d) Liver sampling, biliary diversion, 
and removal of either the upper (c) or the 
lower (d) half of the small intestine: After 
sampling the liver and preparing a biliary 
fistula as in group (a), either the upper or 
the lower half of the small intestine was ex- 



cised. When removing the upper half, the 
duodenum and 1 cm of the proximal 
jejunum were left in place. The continuity 
of the intestine was restored as in proce- 
dure (b) by means of a plastic tube. The 
same was done when removing the lower 
half of the intestine connecting the stump of 
terminal ileum to the remaining upper half 
of the intestine. 

(e) Liver sampling and sham-operation: 
The animals in this group underwent re- 
moval of the lateral portion of the medial 
lobe of the liver followed by manipulation 
of the viscera to mimic the surgical trauma 
of the other groups. No bile fistula was pre- 
pared in this group. 

All animals received a subcutaneous in- 
jection of 20 ml of 5% glucose in saline im- 
mediately after surgery and were main- 
tained at room temperature and fasted for 
the next 24 hr following which they were 
sacrificed and their livers removed for es- 
timation of the rate of cholesterol synthesis. 

Determination of the rate of cholesterol 
synthesis. The samples of liver removed at 
the time of surgery and at the time of sac- 
rifice were washed in ice-cold Tris-HCl 
buffer, 0.1 M pH 7.8, containing 0.01 M 
K2HPO4, 0.03 M nicotinamide, 0.1 mAf 
EDTA, and 0.6 mM MgClj. The pieces 
were then blotted dry, weighed, and 
homogenized in the above buffer (2.5 ml/g 
of liver) containing also 5.0 mAf oxidized 
glutathione. The homogenization was done 
using a loosely fitting glass -Teflon Pot- 
ter-Elvehjem homogenizer and was com- 
pleted in two strokes. The homogenates 
were centrifuged at 500g for 20 min at 
4°. Eight-tenths milliliter of supernatant 
was incubated at 3T in a Dubnoff incubator 
(140 oscillations/min) for 5 min to allow 
temperature equilibration and the synthesis 
was initiated by the addition of 0.2 ml of a 
solution containing 1.8 mAf [2-'^C]sodium 
acetate (specific activity, 0.4 mCi per 
mmole), coenzyme A 50 fiM, ATP 1.0 mAf, 
NADP 0.5 mAf, and glucose 6-phosphate 
3.0 mAf. The incubation was continued for 
30 min at the end of which 2.0 ml of 95% 
ethanol and 0.5 ml of 60% KOH (w/v) werf 
added. After saponification for 90 min at W 
the 3-/3-hydroxy steriods were precipitaU 



186 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



as the digitonides, washed, and counted as 
described before (4). Protein determina- 
tions in the homogenates were done fol- 
lowing the Lowry method with bovine al- 
bumin as standard (5). The results are given 
in terms of nanomoles of acetate incorpo- 
rated into cholesterol per gram of protein 
per hour. The percentage change in cho- 
lesterol synthesizing activity between the 
baseline and the 24-hr post-surgery values 
were calculated for each animal. 

Results. Effect of biliary diversion on the 
rate of hepatic cholesterogenesis 24 hr after 
surgery. Before studying the effect of in- 
testinal removal upon the hepatic choles- 
terol synthetic activity of bile fistula rats, it 
was necessary to demonstrate that the 
stimulation of hepatic cholesterogenesis 
after bile diversion occurred within a period 
short enough to assure the survival of the 
animals with the small intestine removed. 
Penhos et aL (6) demonstrated that eviscer- 
ated rats with a functional liver in which the 
stomach, small and large intestines, pan- 
creas, mesentery and spleen had been ex- 
cised and in which a bile fistula had been 
produced to allow continuity of bile secre- 
tion, survived for more than 72 hr provided 
that dehydration was prevented by the ad- 
ministration of saline. It was therefore safe 
to assume that rats subjected to the much 
less traumatic procedure (b), i.e., liver 
sampling* biliary diversion, and excision of 
the small intestine, would easily survive for 
24 hr. It remained to be demonstrated that 
24 hr after biliary diversion there was al- 
ready a detectable increase in the rate of 
hepatic cholesterogenesis. To this effect 
three rats, fed the maintenance diet until 
the time of the experiment, were subjected 
to procedure (a), i.e., liver sampling and 
biliary diversion, and the rate of hepatic 
cholesterol synthesis was determined at the 
time of surgery and 24 hr later. In these 
animals the incorporation of acetate into 
cholesterol increased significantly from 198 
± 12 (mean ± SE) nmoleg of proteiahr to 
552 r 8 nmole g protein hr, or to 280 ± W'c 
of the baseline value {P ^ O.OOl). 

EtYect of removal of the small intestine 
on the rate of hepatic cholesterol synthesis 
of rats 24 hr after total hiliarx diversion: 



Experiment /. Six animals were subjected 
to procedure (a), liver sampling and biliary 
diversion with intact intestine; five to pro- 
cedure (b), liver sampling, biliary diversion, 
and nearly total removal of the small intes- 
tine; and four to procedure (e), liver sam- 
pling and sham operation. All these animals 
were fed the maintenance diet until the day 
of the experiment. The results are summa- 
rized in Table I. As in the preliminary ex- 
periment, biliary diversion markedly in- 
creased the rate of hepatic cholesterol 
synthesis to 377 ± 65% of the baseline 
value (P < 0.01). Removal of the small in- 
testine in the bile-fistula rats completely 
eliminated this response resulting instead, 
in a small but significant drop of hepatic 
cholesterol synthetic activity to 64 ± 10% 
of the baseline rate (P < 0.01). The sham- 
operated animals showed also a modest but 
significant {P < 0.05) decrease in the rate 
of hepatic cholesterogenesis (Table I). 

Experiment IL To investigate the possi- 
bility that the stimulation seen in the bile- 
diverted rats with intact intestine was due 
to some substance absorbed from the diet, 
the experiment was repeated with animals 
that consumed the maintenance diet until 24 
hr prior to surgery and then given 10% glu- 
cose in water ad libitum as their only food. 
The results are shown in Table I. As in ex- 
periment 1, cholesterol synthesis increased 
very significantly (to 670 ± 79% of baseline, 
P < 0.01) in the bile-diverted rats and de- 
creased slightly (to 60 ± 8% of the baseline 
(P < 0.01) in the bile-diverted animals with 
the jejunum and ileum removed. The 
sham-operated group also showed a small 
decrease in hepatic cholesterogenesis. The 
most striking difference between experi- 
ments 1 and 11 was in the absolute rate of 
cholesterol synthesis which was many 
times higher in the diet-fed than in the 
glucose-fed group. 

Effect of partial removal of the small in- 
testine on the rate of hepatic cholesterol 
synthesis of rats 24 hr after total biliary di- 
version. Experiments 1 and II support the 
existence of a stimulatory factor of hepatic 
cholesterol synthesis produced by the 
jejunum-iieum. In an attempt to determine 
more precisely the portion of the small in- 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



187 



z 

H 
Z 
>• 
CO 



U 

u 

p 

ttJ 
X 
u. 
O 

UJ 



l« 



u 



< X 
CO ^ 



< 
> 
o 



u. 
O 

H 
U 

u 
u. 
u. 
U 



QQ 

< 



^ o 

^ 'C 

w 8 

3 

i 



& 



QQ 









^ 



cS 



'•5 s 

CQ 



CO 2 

a 



+1 +1 



QO 
+1 



*n 00 

•o © 

+1 +1 

— ri 



O 3 — «o 
+1 II -i-l II 



OS 

d 

+1 



•l-l II 
© K 



+1 

3 



5 




+1 


+1 


s 





m >© vd «ri 
il II -i-l II 



il II 



C CO 

b ^ g o- 
-3 .S .S -5 5 I .S 



J ^ .5 I § 5 .S 






1.1 



o O w 






I 



e 

O 

E 






V 
eS 



g 


© 5 


4> 


o-S 


£ 


v^ 


•5 


•i:^ 


73 * 


O 4> 


s :b 


^ C 


rem 
rote 




-o °- 


© — 


§ ^ 


V i 


diversion 
olesterol/g 


0,^ 

11 


biliary 
into ch 


II 


£ -o 


s^t:* 


group wi 

;um. 

corporate 


If 


5 8 = 

w v^ -» 


^ 3 


between 
itz to the 
^CJacetat 


2 E 


-o S>- 


ft o 


gHCi 


a>«t: 


O <^ «fi 


^ V 


« o o . 


C 7k 


ifference 
ligament 
E nmole 
f animals 


2i S 

H 
•II 


-o „co o 


hte. No 
From th< 
Mean ± 
Number 


< 3 * V 


* ++ 



188 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



testine involved in its production, biliary- 
diverted rats with either the upper half or 
the lower half of the small intestine re- 
moved were prepared and the rate of cho- 
lesterol synthesis of their livers before and 
after surgery was compared to that of 
biliary-diverted animals with the small in- 
testine intact. All the animals used in this 
experiment were fed nothing but 10% glu- 
cose ad libitum 12 hr prior to surgery. The 
results are shown in Table II. It can be seen 
that preserving either the upper or the 
lower half of the small intestine was suffi- 
cient to elicit the increase in the rate of he- 
patic cholesterogenesis characteristic of 
bile-fistula rats with intact small intestine 
(Table II). Although the extent of the 
stimulation found in the animals with partial 
removal of the small intestine seemed to be 
lower than in the intact animals, no statisti- 
cally significant difference could be demon- 
strated among the three groups. 

Discussion. It is well established that 
removal of bile from the intestinal lumen, 
by whatever means, results in increased 
rates of hepatic cholesterol synthesis. 
Thus, ligation of the bile duct (7), creation 
of a bile fistula (2), cholestyramine admin- 
istration (8), or ileal bypass (9), all produce 
significant elevations in hepatic choles- 
terogenesis. We have postulated (1) that 
these responses follow a homeostatic 
mechanism, mediated by a stimulatory 



factor produced by the intestine, and serv- 
ing to restore to normal the intraluminal 
concentration of bile acids whenever it 
drops below requirements. The stimulation 
of de novo hepatic bile acids synthesis im- 
plies a stimulation of the biosynthesis of 
cholesterol, their obligatory precursor. This 
hypothesis is supported by the results pre- 
sented in Table I, experiment I. The com- 
plete disappearance of the stimulation of 
liver cholesterogenesis in the bile-fistula 
rats when the jejunum-ileum was removed 
argues strongly for the loss of a stimulatory 
factor or factors contributed by the small 
intestine. 

The presence in portal blood of a sub- 
stance or substances capable of promoting 
higher rates of hepatic cholesterol synthesis 
has been indicated before by the elegant 
experiments of Starzl et al. involving partial 
transposition of portal and vena caval blood 
in dogs (10, 11). In these studies, portal 
blood was supplied to one portal branch of 
the liver while the other portal branch was 
supplied with blood from the inferior vena 
cava. The lobes receiving the portal blood 
had higher cholesterol synthesizing activity 
than the lobes irrigated with systemic 
blood. Starzl and his co-workers attributed 
this difference primarily, but not solely, to 
the much higher concentration of insulin in 
portal than in caval blood. Although the 
role of insulin in maintaining the trophism 



TABLE II. Effect of Partial Removal of the Small Intestine on the Rate of Hepatic 
Cholesterol Synthesis of Rats 24 hr after Total Biliary Diversion 





Baseline rate 


24 hr after surgery 


Surgical procedure 


Rate" 


% of baseline rate 


Biliary diversion with intact intestine 


42 ±6 


258 ± 49* 


615 ± 78t 


(n = 4)^ 








Biliary diversion and removal of the 


45 ± 11 


182 ± 39* 


409 ± 57t 


upper half of the small intestine 








in = 4) 








Biliary diversion and removal of the 


59 ± 10 


303 ± 56t 


511 ±36t 


lower half of the small intestine 








(/; = 5) 









Note. No significant differences found between the three groups of rats either in baseline rates or rates 
after surgery. 
" Mean ± SE nmole of [2-'^C]acetate incorporated into cholesterol/g of protein/hr. 
'' Number of animals. 
* t Significant difference from the baseline rate by paired / test at P < 0.01 (*) or P < 0.001 (t). 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



189 



of the liver cannot be denied, it is difficult 
to attribute our findings to differences in the 
supply of insulin to the liver of the bile- 
diverted animals with and without small 
intestine since in all of our groups the pan- 
creas and its venous drainage into the portal 
vein were preserved intact. Furthermore, 
all animals were fasted after surgery which 
should have reduced the concentration of 
insulin in the portal blood to very low levels 
in all of our groups. Schneider et al, (12) 
have also provided experimental support 
for the existence of a factor in blood capa- 
ble of stimulating liver cholesterol synthe- 
sis. By cross-circulation studies they 
showed that the blood of ileal bypassed rats 
stimulated hepatic cholesterol synthesis in 
their intact parabiosed partners. The ileal 
bypass procedure employed by these au- 
thors diverts the flow of bile from the distal 
half of the small intestine which, according 
to our hypothesis, would then respond by 
producing the postulated stimulatory factor. 

A slight, but significant decrease of he- 
patic cholesterol synthetic activity was ob- 
served in the sham-operated animals and in 
the group with biliary diversion and 
jejunum-ileum removal. This effect is at- 
tributed to the period of fasting (13) after 
surgery. The most important difference, 
however, was the total disappearance of the 
stimulation of hepatic cholesterogenesis in 
the bile-diverted animals with simultaneous 
removal of the jejunum-ileum. These re- 
sults are compatible with the postulate that 
a stimulatory factor is produced by the in- 
testine in response to decreased intralumi- 
nal concentrations of bile. The alternative 
hypothesis, that the intestine generates an 
inhibitor of liver cholesterol synthesis 
which would be produced in lesser quan- 
tities in the bile-diverted animals, can safely 
be discarded since this explanation implies 
a maximal rate of cholesterol synthesis in 
the animals with the small intestine re- 
moved in which production of the inhib- 
itory factor would have dropped to near 
zero. 

Experiment II was an attempt to answer 
the question of whether or not the intestinal 
factor(s) is produced by the intestine or is 



absorbed from the lumen where it may be 
present as a dietary component or as a con- 
stituent of the secretions of the digestive 
apparatus. We reasoned that if a similar 
stimulatory response to that observed in 
experiment I was obtained when the ani- 
mals had received nothing but 10% glucose 
in water for 24 hr prior to surgery, it could 
not be attributed to something absorbed 
from the diet. Although our results show 
comparable percentage increases in the 
rates of cholesterol synthesis of the bile- 
diverted intestine-intact animals in both ex- 
periments I and II, and the total abolition of 
this response by intestinal removal, the 
baseline rates of cholesterogenesis in ex- 
periment II were strikingly lower than those 
of experiment I. We attribute this to the 
effect of food deprivation in experiment II. 
Fasting markedly inhibits the rate of he- 
patic cholesterol synthesis (13) and we have 
calculated that the rats fed 10% glucose as 
their sole food for 24 hr prior to surgery 
(experiment II) consumed during that pe- 
riod less than 10% of the calories ingested 
by the animals fed the maintenance diet 
until the time of surgey (experiment I). 
Furthermore, the rats subjected to partial 
intestinal removal (vide infra) which were 
fed nothing but 10% glucose for 12 hr prior 
to surgery had intermediate baseline rates 
of liver cholesterogenesis, as would be ex- 
pected given their shorter period of food 
restriction. Taken together the results of 
experiments I and II indicate that the 
baseline rate of hepatic cholesterogenesis, 
set as a function of recent food intake, can 
be accelerated about three- to sevenfold fol- 
lowing bile diversion. The intestinal stimu- 
latory factor appears therefore capable of 
altering the baseline rate but is certainly not 
its sole determinant. Furthermore, the re- 
sults of experiment II support the conten- 
tion that the factor is produced by the in- 
testine and not merely absorbed from the 
luminal content. 

Removal of either the upper or lower half 
of the jejunum-ileum failed to prevent the 
stimulation of hepatic cholesterogenesis 
after creation of a bile fistula (Table II). 
These results indicate that the small intestine 



190 



CHOLESTEROGENESIS-INTESTINAL STIMULATION 



has a considerable reserve capacity for pro- 
duction of the stimulatory factor, and that 
this function resides in both the lower and 
upper halves of the jejunum-ileum. It seems 
clear also that the duodenum lacks the ca- 
pacity to produce the stimulatory factor. 
These experiments also indicate that the ab- 
sence of stimulation of cholesterogenesis 
observed in the bile-fistula animals with re- 
moval of the entire jejunum-ileum cannot 
be attributed to the surgical trauma per se. 
There is little difference in the trauma in- 
flicted to the animals undergoing removal of 
half or the whole of the jejunum-ileum. 

Based on the data provided by our ex- 
periments we conclude that removal of bile 
(or more precisely of bile acids) from the 
intestinal lumen leads to the production by 
both the upper and lower halves of the 
jejunum-ileum of a stimulatory factor or 
hormone which reaches the liver via the 
portal circulation and accelerates the rate of 
hepatic cholesterogenesis. The isolation and 
characterization of such a hormone should 
advance our understanding of the physio- 
logical regulatory mechanisms of choles- 
terol homeostasis. 

We gratefully acknowledge the expert technical as- 
sistance of Mr. Darrel Norton. This work was sup- 
ported in part by Veterans Administration and NIH 
Grant 5P01-CA28103; J.E.S. was supported by a fel- 
lowship from the Funda9ao de Amparo e Pesquisa do 
Estado de Sao Paulo, Brazil. 



1. Krumdieck CL, Ho KJ. Intestinal regulation of 
hepatic cholesterol synthesis: an hypothesis. 
Amer J Clin Nutr 30:255-261, 1977. 

2. Myant NB, Eder HA. The effect of biliary drain- 
age upon the synthesis of cholesterol in the liver. 
J Lipid Res 2:363-368, 1%1. 

3. Navia JM, Lopez H, Harris RS. Purified diet for 



dental caries research with rats. J Nutr 
97:133-140, 1%9. 

4. Siperstein MD, Guest MJ. Studies on the site of 
the feedback control of cholesterol synthesis. J 
Clin Invest 39:642-652, 1960. 

5. Lowry OH, Rosenbrough NJ, Farr AL, RandaU 
RJ. Protein measurement with the Folin phenol 
reagent. J Biol Chem 193:265-275. 1951. 

6. Penhos JC, Woodbury C, Tizabi Y, Ramey ER. 
Metabolic studies in eviscerated rats with func- 
tional livers. Proc Soc Exp Biol Med 148:1159- 
1163, 1975. 

7. Fredrickson DS, Loud AV, Hinkelman BT, 
Schneider HS, Frantz Jr ID. The effect of ligation 
of the common bile duct on cholesterol synthesis 
in the rat. J Exp Med 99:43-53, 1954. 

8. Huff JW, Gilfillan JL, Hunt VM. Effect of 
cholestyramine, a bile acid binding polymer on 
plasma cholesterol and fecal bile acid excretion in 
the rat. Proc Soc Exp Biol Med 114:352-355. 
1%3. 

9. Moutafis CD, Myant NB. Increased hepatic syn- 
thesis of cholesterol after ileal by-pass in mon- 
keys. CUn Sci 34:541-548, 1968. 

10. Starzl TE, Lee lY, Porter KA, Putnam CW. The 
influence of portal blood upon lipid metabolism in 
normal and diabetic dogs and baboons. Surg 
Gynecol Obstet 140:381-396. 1975. 

1 1 . Starzl TE, Porter KA, Francavilla JA, Benichou J, 
Putnam CW. A hundred years of the hepato- 
trophic controversy. In: Hepatotrophic Factors. 
Ciba Foundation Symposium 55 (new series). 
Amsterdam, Elsevier, ppll 1-129, 1978. 

12. Schneider PD, Guzman IJ, Rucker RD, Stocks 
TG, Varco RL, Buchwald H. Increased hepatic 
cholesterol synthesis in normal rats by cross- 
circulation with ileal bypassed partners. Athero- 
sclerosis 34:383-389, 1979. 

13. White LW, Rudney H. Regulation of 3-hydroxy- 
3-methylglutarate and mevalonate biosynthesis by 
rat liver homogenates. Effects of fasting, choles- 
terol feeding, and Triton administration. 
Biochemistry 9:2725-2730. 1970. 

Received April 12, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 191-195 (1982) 



Response of the Renal Vitamin D Endocrine System to Oxidized Parathyroid 

Hormone (1-34) (41497) 

ALEXANDER D. KENNY and PETER K. T. PANG 

Department of Pharmacology and Therapeutics, Texas Tech University Health Sciences Center, 

Lubbock, Texas 79430 



Abstract. Two preparations of bovine parathyroid hormone (bPTH), the natural 
bPTH(l-84) and the synthetic bPTH(l-34) fragment, have been treated with hydrogen 
peroxide and assayed for the effect of such treatment on the renal vitamin D endocrine 
system in Japanese quail. The oxidized and untreated preparations were ii\jected intramus- 
cularly into 4- week-old male Japanese quail, 12 hr after which the kidneys were removed 
and homogenized. The kidney homogenates were incubated with tritiated 25- 
hydroxyvitamin D,[25-(OH)D3] and the production rates of l,25-(OH)2D3 and of 24,25- 
(OH),D, were determined as indices of 25-(OH)D3-l -hydroxylase and 25-(OH)D3-24- 
hydroxylase activities, respectively. Both untreated bPTH(l-34) and untreated bPTH(l-84) 
stimulated 1 -hydroxylase and suppressed 24-hydroxylase activities. Oxidation of either 
bPTH(l-34) or bPTH(l-84) did not eliminate these responses, whereas the effects of oxida- 
tion on other responses to bPTH(l-34), namely inactivation of the hypotensive and renal 
adenylate cyclase stimulating responses, were observed as anticipated from our earlier 
observations. The importance of these findings is heightened when viewed in the context of 
our previous reports that oxidation of bPTH(l-34) leaves the hypercalcemic and hypocal- 
ciuric responses intact while partially or possibly totally inactivating all other msgor re- 
sponses studied to date. It may be concluded that the mechanisms involved in effecting the 
hypercalcemic, hypocalciuric, and renal 25-(OH)D3-l -hydroxylase responses to bPTH(l-34) 
demand structural requirements in the peptide molecule which are different from those 
needed to effect the hyperphosphaturic, hypophosphatemic, hypotensive, smooth muscle 
relaxing, and renal adenylate cyclase responses. 



It is well known that the hypercalcemic and 25-hydroxyvitamin D3-24-hydroxylase 
activity of intact bovine parathyroid hor- activities in Japanese quail. The untreated 
mone [bPTH(l-84)] is largely inactivated by and oxidized preparations were also moni- 
mild oxidation with hydrogen peroxide (1, tored for hypercalcemic, hypotensive, and 
2). Recently, using conditions known to renal adenylate cyclase stimulating activities, 
substantially inactivate the hypercalcemic Materials and Methods. Hormone prep- 
activity of bPTH(l-84) in Japanese quail, orations. Synthetic bovine parathyroid 
we have shown that the synthetic fragment, hormone tetratricontapeptide, bPTH(l-34), 
bPTH(l-34), resists such inactivation with was obtained either from Beckman (Beck- 
hydrogen peroxide (3, 4). Further work man Instruments, Inc., Bioproducts Op- 
from our laboratory has revealed that, al- eration, Palo Alto, Calif.; lot No. B01130) 
though the avian hypercalcemic activity of or from Peninsula (Peninsula Laboratories, 
the synthetic bPTH(l-34) is unaffected, the Inc., San Carlos, Calif.; lot No. 001752). 
hypotensive (4), hyperphophaturic (5), and The bPTH(l-34) preparations had assigned 
uterine relaxing (6) activities of bPTH(l-34) potencies of 6000 and 10,000 lU/mg, re- 
are reduced by oxidation with hydrogen spectively, when assayed in vitro using 
peroxide. activation of rat renal cortical adenylate 

We now wish to report the effect of mild cyclase as the response. The intact bovine 

oxidation on the ability of intact bPTH(l-84) parathyroid hormone, bPTH(l-84), was 

and of synthetic bPTH(l-34) to modify the purified material (NIBSC 77/533) kindly 

renal 25-hydroxyvitamin D3-I -hydroxylase supplied by Dr. Joan Zanelli of the National 

191 



192 



OXIDIZED PTH AND VITAMIN D METABOLISM 



Institute for Biological Standards and Con- 
trol, London. It had an assigned potency of 
2300 lU/mg. 

Hydrogen peroxide treatment. One milli- 
liter of a solution of either bPTH(l-84) or 
bPTH(l-34), each dissolved in acid saline 
(0. 154 M NaCI adjusted to pH 3.0 with HCl) 
at a concentration of 1000 lU/ml, was 
treated with 0.3% hydrogen peroxide by 
adding 10 /liI of 30% hydrogen peroxide 
(Fisher Scientific Co.. Pittsburgh, Pa.; 
catalog No. H-325) to the 1.0-ml sample. 
Following addition of hydrogen peroxide, 
the mixture was incubated for 30 min at 25''. 
Control incubations were performed using 
10 /btl of water instead of the 30% hydrogen 
peroxide. The mixture was immediately ly- 
ophilized following incubation. The lyophi- 
lized preparations were sealed and stored 
below 0° until required. They were recon- 
stituted in 1.0 ml of water and diluted in 
acid saline for ii\jection. Total enzymatic 
digestion of the bPTH(l-34) preparation, 
subjected to oxidation under these condi- 
tions, has revealed that such treatment is 
100% effective by reducing the methionine 
content from 1 .74 moles per mole of peptide 
to zero (4). The evidence also indicated that 
the methionine is oxidized mole for mole to 
methionine sulfoxide. No other amino acid 
modiflcutions were detected. 

Renal /- and 24'hydroxylase assays in 
Japanese quail. Four-week-old immature 
male Japanese quail (Coturnix cotumix 
japonica). weighing approximately 100 g 
and bred and maintained in the vivarium of 
the Texas Tech University Health Sciences 
Center as described elsewhere (7), were 
used. The control or bPTH solutions were 
ii\jected intnimuscularly in a volume of 0.4 
ml/bird to five groups of five birds each: (a) 
control, acid saline: (b) bPTH(l-34); (c) 
oxidi/ed bPlH(l-34): (d) bPTH(l-84): and 
(e) o\idi/ed bPTH(l-84). Twelve hours 
after it\ieclion» the birds were sacrificed al 
which lime the kidneys were removed for 
assessment of I- and 24-hydrovylase ac- 
tivities in vitro as described elsewhere (8, 
^). l-ach kidncv homogenale was incubated 
for 10 nun in the pix'sence of 50 pniole of 
25 [>,:?- *H|hvdro\y\itamin D.^ with a 
nccific actiNJtv of ^.6 Ci mmole. The incu- 



bation mixture was extracted with appro- 
priate solvents and the extract was dried, 
dissolved in 300 fil of chlorofonn/hexane 
(65:35), and applied to a 0.7 x 18-cm 
Sephadex LH-20 column. The column was 
eluted with the same solvent system and 50 
fractions (125 drops or 0.92 nil each) were 
collected. The tritium content of each frac- 
tion was determined by liquid scintillation 
counting. Total disintegrations per minute 
(dpm) under each peak was calculated and 
corrected for extraction losses and the data 
were expressed as picomoles min~* g"* kid- 
ney. The l,25-(OH)2D3 peak generated by 
Japanese quail kidney homogenates under 
these circumstances has failed to reveal any 
cochromatographic contaminants when 
subjected to more rigorous analysis (10). 

Hypercalcemic activity in Japanese 
quail. The bPTH preparations were as- 
sayed for hypercalcemic activity in Japa- 
nese quail by the method described by 
Dacke and Kenny (2). All injections were 
given intravenously in a volume of 0.4 ml/ 
bird using an injection vehicle consisting of 
51 mA/ CaCU and 0.01% bovine serum al- 
bumin. 

Renal adenylate cyclase activity. A renal 
membrane preparation of adenylate cyclase 
was made from Japanese quail kidneys and 
used for the in vitro assay of the bPTH pep- 
tides. Details of the protocol, which was 
modified from assays using rat renal corti- 
cal membrane preparations ( 11-13), will be 
supplied on request. The assay is sensitive 
to about 0.01 /xg/tube (2 x 10"** M) of syn- 
thetic bPTH(l-34). 

Hypotensive activity in the rat. The 
hypotensive effect of the bPTH peptides 
was assayed in the rat following intrave- 
nous injection as described elsewhere (14). 
A significant hypotensive response may be 
seen with 1 fig/kg. A typical response to 7 
/Ltg'kg is presented in Fig. 1. 

Plasma calcium. Heparinized blood was 
obtained from the Japanese quail by heart 
puncture following light anesthesia with 
halothane (Fluothane. Ayerst Laboratories, 
Inc., New York. N.V.) and prior to re- 
moNul of the kidneys. Plasma calcium con- 
centration was determined by atomic ab- 
sorption spectrophotometr\ (Perkin- Elmer 



OXIDIZED PTH AND VITAMIN D METABOLISM 



193 



I ito- 



r 1 



I I I I I I I I 

1460101214 



I"- t 



40 HI bmi-(i-a4)pw i« 



I— T — I — r 



Fig. 1. Typical hypotensive response to bPTH(l-34) 
in the anesthetized rat. Upper panel: control injec- 
tion of saline. Lower panel: response to 40 lU/kg of 
bPTH(l-34). 

Model 303) following dilution with 0.5% 
lanthanum chloride and deproteinization 
with 5% trichloroacetic acid. 

Results. Both unoxidized bPTH(l-84) 
and unoxidized bPTH(l-34) significantly 
enhanced renal 1-hydroxylase activity and 



suppressed 24-hydroxylase activity. Mild 
oxidation with hydrogen peroxide did not 
significantly modify these responses (Ta- 
ble I). The hypercalcemic activity of 
bPTH(l-34) was unaffected by oxidation, 
whereas that of bPTH(l-84) was markedly 
reduced (Table II). In contrast, oxidation 
essentially eliminated the hypotensive (Fig. 
2), and significantly reduced the renal 
adenylate cyclase (Table III), activities of 
bPTH(l-34). The plasma calcium concen- 
trations were normal 12 hr following injec- 
tion in all birds as anticipated; the hyper- 
calcemic response to bPTH is very rapid in 
onset and short in duration in Japanese 
quail (15). 

Discussion. The significance of this work 
becomes more apparent by placing it within 
the context of our other structure -activity 
studies associated with hydrogen peroxide 
treatment of bPTH(l-34). Whereas the 
hypercalcemic activity of intact bPTH( 1-84) 
is largely inactivated by mild oxidation with 
hydrogen peroxide, that of the synthetic 
fragment, bPTH(l-34), is unaffected by 
such treatment (3, 4). Prior to our report, 
apparently no laboratory had investigated 
the hypercalcemic activity of oxidized 
synthetic bPTH(l-34). Mild oxidation is as- 
sumed to affect only the methionine resi- 
dues at positions 8 and 18 which occur in 
both bPTH(l-34) and bPTH(l-84). Why, 
then, does such treatment affect the activity 
of one molecule and not the other? Perhaps 



TABLE I. Effect of HjOj on the Response of the Renal 1- and 24-Hydroxylases to 
bPTH(l-34) AND bPTH(l-84) in 4-Week-Oi.d Male Japanese Qhah 











Metabolite production 










(pmolemin 


•g • kidney) 




Dose, im 




Plasma Ca 







Treatment 


(per bird) 


No. birds 


(mg/dl) 


1.25.(OH),D, 


24.25-(OH),D3 


Control, acid saline 


0.4 ml 


5 


9.8 ± 0.24 


2.4 ± 2.35 


27.6 ± 3.44 


bPTH(l-34) 


20Atg 


5 


9.6 ±0.16 


17.2 ± 1.82** 


12.7 ± 3.37* 


bPTH(l-34) -»- HjO, 


20 Mg 


5 


9.4 ±0.21 


22.5 ± 3.57** 


6.4 ± 4.03** 


bPTH(1.84) 


17 Mg 


5 


9.5 ± 0.25 


8.3 ± 3.24*** 


12.4 ±4.25* 


bPTH(l-84) -»- HjO, 


17 Mg 


5 


9.5 ± 0.26 


10.9 ± 1.98* 


8.0 ± 3.30** 



Note. Results are means ± SE. bPTH(l-34): synthetic bovine PTH( 1-34) from Beckman (6000 lU/mg claimed 
potency). bPTH(l-84): purified bPTH(l-84) from National Institute for Biological Standards and Control. 
London (NIBSC 77/533) with an assigned potency of 2300 lU/mg. 
* Significant at P < 0.05. 
•• Significant at P < 0.01. 
*•* Significant slX P < 0.10 (one-tail test). 



194 



OXIDIZED PTH AND VITAMIN D METABOLISM 



TABLE II. Effect of HjOj on the Hypercalcemic Response to bPTH(l-34) and 
bPTH(l-84) IN Japanese Quail 



Treatment 



Dose, iv 
(per bird) 



No. birds 



Plasma Ca 
(mg/dl, mean ± SE) 



Assay PI 20 
Control, vehicle alone 
bPTH(l-34) 
HjO^treated bPTH(l-34) 

Assay P162° 
Control, vehicle alone 
bPTH(l-84) 
H,Oj-treated bPTH(l-84) 



0.4 ml 
7/Ltg 
7/Ltg 



0.4 ml 
44/Ltg 
44/Ltg 



10.1 ±0.28 

12.2 ± 0.34*** 
13.0±0.16*** 



9.4 ±0.21 
12.4 ±0.31*** 
10.2 ± 0.52 



Note. See footnotes to Table I. Vehicle: 51 mA/ CaCls + 0.01% bovine serum albumin. 
" These data (assay PI 62) are reproduced from Pang et al. (4). 



oxidation of the two methionines at posi- 
tions 8 and 18, respectively, results in con- 
formational changes such that the intact 
bPTH(l-84) molecule, with its large biologi- 
cally inactive C-terminal tail, cannot ade- 
quately interact with its receptor. The syn- 
thetic N-terminal fragment, bPTH(l-34), on 



DOSE OF bPTH(l-34), unit»/kg 



® ©- -« 

bPTH{l-34) ♦ H2O2 



bPTH(|-34) 




i 

"i 



Fig. 2. Effect of different doses of untreated and 
oxidized bPTH(l-34) on mean arterial pressure in the 
anesthetized rat. 



the other hand, may not experience any 
steric hindrance resulting from such a con- 
formational change. 

Equally important is our finding that mild 
oxidation partially or possibly totally inac- 
tivates some but not all of the responses 
associated with the synthetic bPTH(l-34) 
fragment. Of the responses studied to date, 
three resist inactivation. These are: (i) the 
hypercalcemic response in the Japanese 
quail (3, 4); (ii) the hypocalciuric response 
in the rat (16): and now, from the present 
study, (iii) the ability to activate the renal 
25-hydroxy vitamin Dg-l-hydroxylase and 
suppress the 25-hydroxyvitamin D3-24- 
hydroxylase enzymes in the Japanese quail. 
All three responses are classically as- 
sociated with the mobilization of calcium 
into extracellular fluid from bone, kidney, 
and gut. Other responses studied are par- 
tially or totally eliminated by oxidation of 
bPTH(l-34). These include: (i) the hyper- 
phosphaturic response in the rat (5); (ii) 
renal adenylate cyclase activation in the 
Japanese quail (Table III); (iii) relaxation of 
induced contraction of smooth muscles (6); 
and (iv) the hypotensive response in the rat 
((4) and Fig. 2). 

One other interesting point emerges from 
these findings. It is obvious that the mecha- 
nisms involved in the hypercalcemic, 
hypocalciuric, and renal 1 -hydroxylase re- 
sponses demand structural requirements in 
the bPTH(l-34) molecule which are differ- 



OXIDIZED PTH AND VITAMIN D METABOLISM 



195 



TABLE IIL Effect of H,Ot on the Avian Renal Adenylate Cyclase Response to bPTH(l-34) 



Treatment 



Dose 
(per tube) 



Adenylate cyclase activity 
(pmole cAMP/mg protein) 



Control 
bPTH(l-34) 
bPTH(!-34) + H,0, 



0.1 Mg 
0.1 Mg 



101(87-115) 

408(361-454) 

131(121-141) 



Note. Results are means; range of duplicates in parentheses. bPTH(l-34): synthetic bPTH(l-34) from Penin- 
sula (10,000 lU/mg). 



cnt from those needed to effect the re- 
sponses, such as the hyperphosphaturic, 
renal adenylate cyclase, hypotensive, and 
smooth muscle responses, which are not 
elicited by the oxidized form of bPTH(l-34). 
Is it possible that the difTerent structural re- 
quirements revealed in our functional stud- 
ies using the oxidation technique approach 
are related to the existence of different 
types of receptors? Further work is needed 
before this question can be answered with 
assurance. 

The authors are grateful for the technical assistance 
of Ingrid M. Greene and Jong-Chaur Shieh. This work 
was supported in part by NIH Grants AM 19475, AM 
21822, and American Heart Association Grant 81794. 



1. Tashjian AH, Ontjes DA, Munson PL. Alkylation 
and oxidation of methionine in bovine parathyroid 
hormooe: Effects on hormonal activity and an- 
tigenicity. Biochemistry 3:1175-1182, 1964. 

2. Dacke CD, Kenny AD. Avian bioassay method 
for parathyroid hormone. Endocrinology 92: 
463-470, 1973. 

3. Kenny AD, Fsmg PKT. Failure of bPTH(l-34) to 
l>e inactivated by oxidation with hydrogen 
peroxide. Calcif Tissue Int 31:73, 1980. 

4. Fsmg PKT, Yang MCM, Keutmann HT, Kenny 
AD. Structure activity relationship of parathyroid 
hormooe: separation of the hypotensive and the 
hypercalcemic properties. Endocrinology, in 
press. 

5. Kenny AD, Pang PKT. Dissociation of the hy- 
percalcemic and hyperphosphaturic actions of 
bovine parathyroid hormone (1-34). Fed Proc 
40:399. 1981. 

6. Shew RL. Yee JA, Fang PKT, Kenny AD. Direct 
effect of parathyroid hormone on rat uterine con- 
traction. Anat Rec 199:234A, 1981. 



7. Baksi SN, Kenny AD. Vitamin D, metabolism in 
immature Japanese quail: Effects of ovarian hor- 
mones. Endocrinology 101:1216-1220, 1977. 

8. Kenny AD. Vitamin D metabolism: Physiological 
regulation in egg-laying Japanese quail. Amer J 
Physiol 230:1609-1615, 1976. 

9. Baksi SN, Kenny AD. Effects of administration of 
antiestrogen (tamoxifen) in vivo on the metabo- 
lism of 25-hydroxyvitamin D, in vitro in the 
Japanese quail. Biochem Pharmacol 26:2439- 
2443, 1977. 

10. Baksi SN, Kenny AD. Vitamin D metabolism in 
aged Japanese quail: Dietary calcium and estrogen 
effects. Amer J Physiol 241:E275-E280, 1981. 

11. Marcus R, Aurbach GD. Bioassay of parathyroid 
hormone in vitro with a stable preparation of 
adenyl cyclase from rat kidney. Endocrinology 
85:801-810, 1969. 

12. White AA, Zenser TV. Separation of cyclic 3',5'- 
nucleoside monophosphates from other nucleo- 
tides on aluminum oxide columns. Application to 
the assay of adenyl cyclase and guanyl cyclase. 
Anal Biochem 41:372-396, 1971. 

13. Forte LR, Nickols GA, Anast CA. Renal adenyl- 
ate cyclase and the interrelationship between 
parathyroid hormone and vitamin D in the regula- 
tion of urinary phosphate and adenosine cyclic 
3',5'-monophosphate excretion. J Clin Invest 
57:559-568, 1976. 

14. Pang PKT, Tenner Jr TE. Yee JA, Yang M, 
Janssen HF. Hypotensive action of parathyroid 
hormone preparations on rats and dogs. Proc Nat 
Acad Sci USA 77:675-678, 1980. 

15. Kenny AD. Dacke CD. The hypercalcaemic re- 
sponse to parathyroid hormone in Japanese quail. 
J Endocrinol 62:51-53, 1974. 

16. Kenny AD, Pang PKT. Effect of oxidation on the 
urinary calcium response in rats to bovine 
parathyroid hormone. Calcif Tissue Int, in press. 



Received November 16. 1981. 
171. 



P.S.E.B.M. 1982, Vol. 



198 



a2-ADRENERGIC RECEPTORS ON PANCREACTIC ISLETS 



U 



< 

z 



Q 

Z 
< 

o 

Ij 
o 





o 

(« 
u 

z 

Ul 

0^ 

O 
< 

« 

Q 

Z 
< 



H 
Z 
u 

Hi 

U 

< 

0. 



Q 

z 
< 

G 
Z 

Q 

Z 

QQ 



QQ 

< 






j2 

a 



CQ 






23 



o 



•c 

•a® 



o 
Z 



SCO 



.5 S 



^'1 



00^ 

E.S 

is 



11 


2g 


do 


+1 +1 


SI 


do 


r- ON 


— ri 


+1 +1 


q OS 


>S OS 


2 i2 
o o 


dc> 


+1 +1 


fs 


d d 


So 


dd 


+1 +1 


«nSo 


<6<6 


S:: 


dd 


+1 +1 




d d 




rn -^ 


+1 +1 


"* O 


d ^ 


«^ 


j-Clonidine (5) 
Dihydroergocryptine 


1 3 « 1 



» 


4> 




c 


5 


l 


4> 


c 


O 


o 


C 


7> 






c 


> 


ed 


w 


o 


o 


C 


£ 


1 


v 


• mm 




f/i 


tf) 


s 


i 


2 


73 




4> 


i ^ 


^ 


a 
'"6 


B 




§ 




>< 


S) 






'S 


"S 




o 


e 


•5 


o 


"2 




2 


5< 


j= 




*rf 


vi 


o 


73 n 


E 


M 


1 




c 


1 


c 


'•o 


o 


& 


§ 


M 



"O -o 


as 


§ 


^ 





, ■* 


4> 


5< 


£ 


"g 


4) 




(5 






M 


"E 


1 


M 


CQ 


^ 


<^ 


o 


o 




c 
.2 


1 


1 


.b 

v 


v 


£ 


& 


c 




o 


S 


S 


u 


JZ 


CO 


a 




V 


+ 


c 


(/} 


a 


§ 


c 




•o 


?> 


c 




e« 


<A 


c 


"3 


■c 


•/) 


x: 


u 


a 


OS 


a> 




-LOa [ ANTAGONIST ]fM) 

Fig. 3. Displacement curves for yohimbine (•) and 
prazosin (O) vs 10 nA/ (^KJclonidtne; yohimbine (of 
antagonist) was; found to displace the bound clonidine 
with an affinity (^f) similar to thai for displacing 
['H]dihydroei^ocryptine. Prazosin (cr, -antagonist) was 
inefTective suggesting that a-receptors on the pancre- 
atic islet cells are of the a, type. 



Fig. 4. The Kf for yohimbine was 7.1 ± 75 
nAf , again indicating the preferential di$*« 
placement at the a^-adrenergic receptor 

The use of adrenergic receptor agonists 
and antagonists in vivo and in vitro has 
led to the generally accepted working hy- 
pothesis that a-adrenergic receptors on 




10 t • 7 • 5 
•LOO [antagonist ](M) 

Fig. 4. Displacement curves for yohimbine (•) and 
prazosin (O) vs 10 nA/ [^H]dihydroergocryptine. 
Yohimbine displaced the bound radioligand from a 
Mngic population of a-receptors whereas prazosin was 
ineffective. The similarity of the data for the two 
radioligands (also in Table 1) indicates that pH]dihy- 
droergocryptine binds to aj-receptors in the islets. 



a2-ADRENERGIC RECEPTORS ON PANCREACTIC ISLETS 



199 



pancreatic )3 cells inhibit insulin secretion 
whereas the )3-adrenergic receptors stimu- 
late it. With the availability of more specific 
adrenergic agonists and antagonists it has 
been suggested that the effects on insulin 
secreting responses are exerted by discrete 
subtypes of receptor population. Thus clon- 
idine, which is an a2-receptor agonist, has 
been shown to inhibit insulin secretion (7) 
and this response is blocked most effective- 
ly by the aj-antagonists, e.g., yohimbine, 
and not by ai-antagonists, e.g., pra- 
zosin (8). 

The present study provides more direct 
measurements of adrenergic receptor den- 
sity and affinity by using radioligands. 
Clearly, the data on both the displacement 
of [^H]clonidine by the various agonists 
(Fig. 2) and by the two specific antagonists 
(Fig. 3) support the conclusion that the a- 
adrenergic receptors on the rat pancreatic 
islets are of the as subtype. 

The mechanisms whereby the various 
types of adrenergic receptors exert their 
effect on insulin secretion are unknown. 
Some suggestions have been made regard- 
ing their role in adipose tissues metabo- 
lism (9). In this tissue activation of /3-ad- 
renergic receptors stimulates adenylate cy- 
clase activity and activation of aj-receptors 
is believed to counter this stimulation. The 
general applicability of these mechanisms 
remains uncertain inasmuch as aj-re- 
ceptors have been reported only in human 
and hamster fat cells (10) but not in rat 
adipocytes (11). 

Attempts to implicate cyclic AMP in the 
a-adrenergic inhibition of insulin secretion 
in rat pancreatic islets have given con- 
tradictory results. Although addition of 
clonidine to such cells clearly inhibited 
glucose-induced insulin secretion, there 
was no afTect on adenylate cyclase activity 
(12). By contrast, it has also been reported 
that clonidine, epinephrine, and norepi- 
nephrine were effective in diminishing the 
glucose-induced accumulation of cyclic 
AMP in islet cells (8). The reason for the 
discrepant data is not clear. 

It is of interest that glucose per se in- 
creases islet adenylate cyclase activity and 
results in accumulation of cyclic AMP in 



the islets. Likewise, stimulation of /3- 
adrenergic receptors increases the adeny- 
late cyclase and accumulation of cyclic 
AMP. Although the /3-adrenergic receptor 
blocker, propranolol, decreases insulin se- 
cretion evoked by a glucose load (13) it is 
not known if there is a link between )3- 
adrenergic receptors and insulin secretion 
evoked by glucose or whether the /3- 
adrenergic system participates at some later 
stage in insulin secretion. Also it is con- 
ceivable that propranolol inhibited insulin 
secretion through its local anesthetic effect 
which is independent of its adrenergic 
blocking action. 

Last, it has been reported, in abstract 
form, that addition of epinephrine to islets 
increased insulin secretion when glucose 
was absent from the media, but it decreased 
insulin secretion in the presence of glucose 
(14). We have found that presence or ab- 
sence of glucose in the media has no effect 
on the a- or /3-adrenergic receptor density 
or affinity (1,2). Thus it appears likely that 
the factors which determine whether the 
adrenergic influence will increase or de- 
crease insulin secretion are at sites beyond 
the adrenergic receptors. 



1. Cherksey B, Altszuler N, Zadunaisky, J. Prepon- 
derance of /3-adrenergic binding sites in pancreatic 
islet cells of the rat. Diabetes 30:172-174, 1981. 

2. Cherksey B, Mendelsohn S, Zadunaisky J, 
Altszuler N. Displacement of a- and /3- 
radioligands by specific adrenergic agonists in rat 
pancreatic islets. Submitted for publication, 1982. 

3. Lacy PE, Kostianovsky MM. Method for the iso- 
lation of intact islets of Langerhans from the rat 
pancreas. Diabetes 16:35-39, 1%7. 

4. Cherksey B, Zadunaisky J, Murphy RB. Cyto- 
skeletal constraint of the /S-adrenergic receptor in 
frog erythrocyte membranes. Proc Nat Acad Sci 
USA 77:6401-6405, 1980. 

5. Scatchard G. The attractions of proteins for small 
molecules and ions. Ann NY Acad Sci 51:660- 
672, 1949. 

6. Cheng Y, Pnisoff WH. Relationship between the 
inhibition constant {Kf) and the concentration of 
inhibition which causes 50 percent inhibition (Iso) 
of an enzyme reaction. Biochem. Pharmacol. 
22:3099-3108, 1973. 

7. Metz SA, Halter JB. Robertson RP. Induction of 
defective insulin secretion and impaired glucose 



200 



a2-ADRENERGIC RECEPTORS ON PANCREACTIC ISLETS 



tolerance by clonidine. Diabetes 27:554-562, 
1978. 

8. Yamazaki S, Katada T, Ui M. Alphas-adrenergic 
inhibition of insulin secretion via interference with 
cyclic AMP generation in rat pancreatic islets. 
Mol Pharmacol 21:648-653. 1982. 

9. Garcia-Sainz J A, Fain JN. Regulation of adipose 
tissue metabolism by catecholamines: Roles of 
alpha 1, alphax, and beta-adrenoreceptors. Trends 
Pharmacol Sci (TIPS) 3:201-203. 1982. 

10, Bums TW. Langley PE, Terry BE, Bylund DB, 
Hoffman BB. Tharp MD, Lefkowitz RJ, Garcia- 
Sainz JA, Fain JN. Pharmacological characteri- 
zation of adrenergic receptors in human adipo- 
cytes. J Clin Invest 67:467-475, 1981, 

11. Fain JN. Garcia-Sainz JA. Role of phos- 
phatidylinositol turnover in alpha^ and of adeny- 
late cyclase inhibition in alphas effects of cate- 
cholamines. Life Sci 26:1183-1194, 1980. 



12. Leclercq- Meyer V, Herchuelz A, Valvcrdc I, 
Couturier E, Marchand J, Malaisse WJ. Mode of 
action of clonidine upon islet function. Dis- 
sociated effects upon the time course and mag- 
nitude of insulin release. Diabetes 29:193-200, 
1980. 

13. Cerasi E, Luft R, Efendic S. Effect of adrenergic 
blocking agents on insulin response to glucose 
infusion in man. Acta Endocrinol 69:335-346. 
1972. 

14. Wollheim CB, Sharp GWG. Stimulatory and in- 
hibitory effects of epinephrine on islet Ca*"" up- 
take and insulin release. Diabetologia 15:282, 1982 
(Abstract). 



Received June 21. 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 201-206 (1982) 

Myoelectric Activity of the Diverted Antroduodenum in the Dog^ (41499) 
J. RUSSELL, P. BASS,2 and A. MIYAUCHI 

School of Pharmacy, Center for the Health Sciences, University of Wisconsin -Madison, 

Madison, Wisconsin 53706 



Abstract. Electrodes were chronically implanted on the gastric antrum and the orad 
duodenum of four dogs. After implantation, fasted and fed state myoelectric activity was 
monitored. After control testing, the stomach was bisected at the orad antral margin and the 
antrum was closed. Bowel continuity was restored with gastrojejunostomy. Extrinsic nerves 
of the antroduodenal segment were maintained. After surgery, fasted and fed state 
myoelectric activity were reassessed. Surgical diversion of the antroduodenum was previ- 
ously shown to result in chronic hypergastrinemia of a postprandial magnitude. It also 
resulted in separation of the antrum from its myoelectric pacemaker in the orad corpus, and 
a chronic absence of digesta from the diverted segment. In the present experiments both the 
antrum and duodenum exhibited unchanged migrating myoelectric complex periods. In ad- 
dition, the duodenal BER frequency was unchanged. In contrast, the antral BER frequency 
was reduced by half. The percentage of antral and duodenal BER superimposed with spike 
potentials in response to liquid or solid meals was unchanged postoperatively despite the 
absence of food in the diverted segment, and despite the absence of postprandially elevated 
gastrin. We conclude that a hormone other than gastrin, and/or a neurally conducted impulse 
generated by the physical presence of the food in the gastric remnant, contributed to the 
postprandial generation of spike potentials in the diverted antroduodenum. 



Distinct electric and motor patterns have 
been characterized in dogs in both the 
fasted and fed states (1,2). Meals of liquid 
and solid foods elicit a characteristic an- 
troduodenal myoelectric pattern where one 
or two duodenal spike potentials immedi- 
ately follow the antral BER and spike po- 
tential (3, 4). This well-timed spike poten- 
tial activity corresponds to antral and 
duodenal contractions that coordinate gas- 
tric emptying. Emptying probably occurs as 
a result of an antral contraction pushing 
chyme into a relaxed duodenum, immedi- 
ately followed by duodenal contractions 
that propel the gastric effluent caudally (5, 
6). Because this electric pattern exists 
postprandially only, the physical presence 
of food seems prerequisite to pattern devel- 
opment (3). 

We have recently reported the use of an 
experimental model where the stomach was 
bisected, the orad antrum closed, and tract 



» Supported by NIH Grant AM15417. 

* To whom all correspondence should be addressed. 



continuity restored via gastrojejunostomy 
(antroduodenal exclusion) (7). This proce- 
dure resulted in chronic diversion of ingesta 
from the antrum and duodenum, and a 
chronic hypergastrinemia which was not 
further elevated by feeding. The postpran- 
dial elevation of gastrin has long been pos- 
tulated to play a major role in generating 
gastric (8, 9) and intestinal (10) spike po- 
tentials and has tacitly been assigned a role 
in the generation of postprandial electric 
patterns. In this study we tested whether 
the diverted antroduodenum could still ex- 
hibit the usual postprandial electric pattern, 
despite no direct contact with food, and an 
absence of postprandially elevated gastrin. 
We report here that the diverted an- 
troduodenum did exhibit a significant, 
though reduced, amount of coordinated 
electric activity postprandially. We suggest 
that an extrinsic neural reflex and/or a non- 
gastrin hormonal efTect may have initiated 
this activity. 

Materials and Methods. Surgical mod- 
eL The surgical preparation of this model 
in our laboratory has recently been de- 



201 
AU hghu reserved. 



202 



CANINE MYOELECTRIC ACTIVITY 



scribed (7). Briefly, four dogs of mixed 
breed (10-20 kg) were anesthetized with 
pentobarbital, 30 mg/kg intravenously (Ab- 
bott Lab). Silver monopolar electrodes 
whose construction and method of implan- 
tation have been reported (4) were im- 
planted on the gut. One electrode was su- 
tured to the serosa of the gastric antrum ca. 
2-4 cm orad to the gastroduodenal junc- 
tion, and two electrodes were implanted on 
the duodenum ca. 2 and 4 cm caudad to the 
gastroduodenal junction. A reference elec- 
trode was sewn into the subcutaneous tis- 
sue of the left flank. A 2-week recovery pe- 
riod was allowed after which control testing 
was performed. After testing, the animals 
were again prepared for surgery. At ce- 
liotomy, the stomach and attending neuro- 
vascular processes were transected at the 
gastric incisura. Care was taken to pre- 
serve the nerve and vascular supply to the 
antrum. The antral margin was then closed, 
removing the gastric antrum from its elec- 
tric pacemaker in the orad corpus. The 
gastric remnant was partially closed and 
anastomosed end-to-side to the jejunum ca. 
10 cm caudad to the ligament of Treitz 
causing antroduodenal exclusion from the 
normal digestive path (Fig. 1). After a 
two- week recovery the dogs were retested. 
In all cases an 18-hr fast preceded the tests. 

Tests. Several myoelectric parameters in 
the interdigestive and digestive states were 
compared pre- and postoperatively to 
characterize the motility of the diverted an- 
troduodenal segment. 

Interdigestive activity. Antral and duode- 
nal interdigestive migrating myoelectric 
complexes (MMC) were identified. The in- 
terdigestive state is marked by a continuum 
of myoelectric activity phases where 
phases 1, 2, and 3 represent intervals of 
low, intermediate, and high numbers of 
spike potentials, respectively, superim- 
posed upon the basic electric rhythm (BER) 
of the stomach and small bowel (1). The 
periodicity of the MMC (time interval from 
phase 3 to the subsequent phase 3, usually 
ca. 100 min) was measured for the antrum 
and duodenum and each were compared 
pre- and postoperatively. Both antral and 
duodenal BER frequencies were calculated 




Fig. 1. Surgical diversion of the gastric antrum and 
duodenum. (A) Antrum, (B) gastric body. (D) 
duodenum, (J) jejunum, (e) electrode. 



from samples taken from the first 30 min of 
phase 1 . Postoperative interdigestive antral 
records were further examined qualitatively 
for the presence of antral BER cycles that 
were of greater than normal frequency (>5 
BER/min), or for the presence of duodenal 
BER superimposed upon the antral BER. 

Digestive state activity. Antral and 
duodenal spike potential responses to both 
liquid and solid meals were each compared 
pre- and postoperatively. The composition 
and gastric emptying properties of the liq- 
uid meal have been described (11). At the 
beginning of phase 1 of the interdigestive 
state, 300 ml of a citrate -fat liquid test 
meal was fed via an oral-gastric tube. Test 
meals were administered during phase 1 to 
prevent confounding of the postprandial 
myoelectric response by spontaneous (i.e., 
phase 2 or 3) spike potential activity. The 
percentage of antral and duodenal BER 
superimposed with spike potentials in re- 
sponse to the meal was determined for the 
30 min immediately following feeding. This 
method was also used to determine the re- 
sponse to 250 g of canned dog food (Vets, 
Perk Foods Co. Inc.). However, after 
feeding of solid food, the response was 
quantified from 15 to 45 min. 

In addition to the presence of duodenal 
spike potentials, their distribution relative 



CANINE MYOELECTRIC ACTIVITY 



203 



to the antral BER cycle was examined. In 
unoperated dogs, postprandial spike poten- 
tials are superimposed on the first and/or 
second duodenal BER that immediately 
follows the antral BER (3). Because the 
ratio of duodenal to antral cycles is ca. 
4.5: 1, the duodenal spike potentials usually 
appear concurrently with the first half of 
the antral cycle. This electric relationship 
was considered to be retained postoperatively 
if the duodenal spike potentials occurred 
only during the first half of the antral cycle. 
The percentage of antral cycles so accom- 
panied by duodenal spike potentials was 
determined pre- and postoperatively after 
both liquid and solid meals. 

Daia analysis. Control and postoperative 
values obtained during the digestive and 
interdigestive states were each compared 
using the / test for unpaired values. In all 
cases, values used in the comparisons rep- 
resent the mean of one to three observa- 
tions per dog. 

Results. Interdigestive state. Surgical 
separation of the gastric antrum from its 
pacemaker in the orad corpus resulted in a 
reduced antral BER frequency (bradygas- 
tria, Table I), and the cycles were ar- 
rhythmic (Fig. 2, A vs B and C). In con- 
trast, the duodenal BER frequency was un- 
altered postoperatively (Table I). There 
was no evidence that the duodenal pace- 
maker hastened the frequency of the un- 
paced, slowed antral BER, and there was 
no superimposition of duodenal rhythms 
upon the antral record. Neither the di- 
verted antrum nor the duodenum exhibited 



changes in the MMC period postoperatively 
(Table I). 

Digestive state. Feeding abolished mi- 
grating myoelectric complexes on the di- 
verted segment. Both the antrum and the 
duodenum exhibited fed state electric pat- 
terns after meals of liquid and solid despite 
being removed from the digestive path. 

Preoperatively, the usual temporal re- 
lationship between postprandial duodenal 
spike potentials and the antral BER cycle, 
as first described by Allen et ai. (3) was 
confirmed. In contrast, after antroduodenal 
exclusion, postprandial duodenal spike 
potentials either accompanied the first half 
of the arrhythmic antral cycle (Fig. 2B), or 
occurred throughout the duration of the 
antral cycle (Fig. 2C). These two post- 
operative patterns appeared with equal fre- 
quency after both liquid and solid meals. 

Antroduodenal exclusion did not aftect 
the overall percentage of antral or duodenal 
BER which were associated with spike po- 
tentials in response to liquid and solid meals 
(Table II). However, the distribution of 
postprandial duodenal spike potentials was 
changed postoperatively. Preoperatively, 
after solid food, 80 ± 6% of the antral cy- 
cles were accompanied by duodenal spike 
potentials which appeared on the first 
and/or second BER immediately following 
the antral cycle (Fig. 2A). Postoperatively, 
after solid food, 41 ± 6% of the antral cy- 
cles remained temporally associated with 
duodenal spike potentials. In contrast, after 
the liquid meal, the degree of temporally 
related response was not changed post- 



TABLE I. Interdigestive Basic Electric Rhythm (BER) Frequencies and Migrating Myoelectric 
Complex (MMC) Periods Before and After Antroduodenal Exclusion 





Organ 


Values 






Parameter 


Preop 


Postop 


pa 


BER frequency (cpm) 
MMC period (min) 


Antrum 
Duodenum 

Antrum 
Duodenum 


5.1 ±0.1M3)*- 
18.8 ± 0.7 (3) 

110 ± 18(4) 

111 ±7(4) 


2.2 ± 0.4 (3) 
18.2 ± 0.3 (3) 

105 ± 9 (3) 
97 ± 6 (3) 


<0.005 
>0.05 

>0.05 
>0.05 


" t test for unpaired values. 
* Mean ± SEM. 
<* Number of dogs. 











206 



CANINE MYOELECTRIC ACTIVITY 



pus, without damaging the gastroduodenal 
junction. Postoperative bradygastric signals 
remained independent of the duodenal 
pacemaker influence since no evidence of 
antral tachygastria (>S cpm) was obtained. 
This confirms the presence of a functional 
insulatory zone and supports the conclu- 
sion (19) that transpyloric BER conduction 
does not influence antral or duodenal BER 
frequencies. 

1. Code CF, Mariett J A. The interdigestive myo- 
electric complex of the stomach and small bowel 
of dogs. J Physiol 246:289-309, 1975. 

2. Carlson GM. Ruddon RW, Hug CC Jr. Bass P. 
Effects of nicotine on gastric antral and duodenal 
contractile activity in the dog. J Pharmacol Exp 
Ther 172:367-376, 1970. 

3. Allen GL, Poole EW, Code CF. Relationships 
between electrical activities of antrum and 
duodenum. Amer J Physiol 207:906-910, 1964. 

4. McCoy EJ, Bass P. Chronic electrical activity of 
gastroduodenal area: Effects of food and certain 
catecholamines. Amer J Physiol 205:439-445, 
1%3. 

5. Carison HC, Code CF, Nelson RA. Motor ac- 
tion of the canine gastroduodenal junction: A 
cineradiographic, pressure, and electric study. 
Amer J Dig Dis 11:155-172, 1966. 

6. Bass P, Russell J. Gastric emptying of liquids: 
Role of the small intestine. In: Chey WY, ed. 
Functional Disorders of the Digestive Tract. 
Raven Press, New York, in press, 1983. 

7. Russell J, Bass P, Shimizu M, Miyauchi A, Go 
VLW. Canine intestinal ulcer: Myoelectric com- 
ponents and the effect of chronic hypergas- 
irinemia. Gastroenterology 82:746-752, 1982. 

8. Kelly KA. Effect of gastrin on gastric myoelectric 
activity. Amer J Dig Dis 15:399-405, 1970. 

9. Thomas PA, Schang J, Kelly KA, Go VLW. 
Can endogenous gastrin inhibit canine interdiges- 
tive gastric motility? Gastroenterology 78:716- 
721, 1980. 



10. Weisbrodt NW, Copeland EM, Keaiiey RW. 
Moore EP, Johnson LR. Effects of pentagastrin 
on electrical activity of small intestine of the dog. 
Amer J Physiol 227:425-429, 1974. 

11. Weisbrodt NW, Wiley JN, Overholt BF, Bass P. 
A relation between gastroduodenal muscle con- 
tractions and gastric emptying. Gut 10:543-548, 
1969. 

12. Thomas PA, Kelly KA. Hormonal control of in- 
terdigestive motor cycles of canine proximal 
stomach. Amer J Physiol 237:E192-E197, 1979. 

13. Eeckhout C, De Wever I, Pecters T, Hellemans J, 
Vantrappen G. Role of gastrin and insulin in post- 
prandial disruption of migrating complex in dogs. 
Amer J Physiol 235:E666-E669. 1978. 

14. Bueno L, Ruckebusch M. Insulin and jejunal 
electrical activity in dogs and sheep. Amer J 
Physiol 230:1538-1544, 1976. 

15. Mukhopadhyay AK. Thor PJ, Copeland EM, 
Johnson LR, Weisbrodt NW. Effect of cholecys- 
tokinin on myoelectric activity of small bowel of 
the dog. Amer J Physiol 232:E44-E47, 1977. 

16. Mukhopadhyay AK, Johnson LR, Copeland EM, 
Weisbrodt NW. Effect of secretin on electrical 
activity of small intestine. Amer J Physiol 
229:484-488, 1975. 

17. Engstrom ER Jr, Webster JG, Bass P. Analysis of 
duodenal contractility in the unanesthetized dog. 
IEEE Trans Biomed Eng BME-26:517-523, 
1978. 

18. Bass P, Code CF, Lambert EH. Electric activity 
of gastroduodenal junction. Amer J Physiol 
201:587-592, I%1. 

19. Gladen HE. Kelly KA. Independence of canine 
gastric and duodenal pacesetter potentials shown 
by electric pacing. Mayo Clinic Proc 52:51-53, 
1977. 

20. Gullikson GW, Okuda H, Shimizu M, Bass P. 
Electrical arrhythmias in gastric antrum of the 
dog. Amer J Physiol 239:G59-G68, 1980. 



Received March 10, 1982. P.S.E.B.M. 1982. Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 207-213 (1982) 



Ovulation, Ovarian 17a-Hydroxylase Activity, and Serum Concentrations of 

Luteinizing Hormone, Estradiol, and Progesterone in Immature Rats with 

Diabetes Mellitus Induced by Streptozotocin (41500) 



MARY S. VOMACHKA and DONALD C. JOHNSON* 

Departments of Physiology and Obstetrics A Gynecology, The Ralph L. Smith Research Center, 
University of Kansas Medical Center, Kansas City, Kansas 66103 



The 



Abstract. Immature female rats were injected with streptozotocin (60 mg/kg, iv) 3 to 4 
days prior to the iiyection (iv) of 20 lU of pregnant mare's serum gonadotropin (PMS). 
Animals were killed at various intervals and the serum levels of estradiol, luteinizing hor- 
mone (LH), and progesterone were determined by radioimmunoassays. The ovarian steroid 
17a-hydroxylase activity was determined by a tritium exchange assay using pregnenolone as 
the substrate. Ovulation was determined 72 hr after PMS by flushing of the oviducts. The 
diabetes mellitus induced by the drug reduced the number of animals ovulating and in some 
animals the number of ova shed when compared to controls. However, a surge in LH, which 
reached a peak at 60 hr, was seen in the diabetic animals; a larger peak with the same timing 
was found in the controls. Changes in ovarian 17a-hydroxylase also indicated that an in- 
crease in LH release occurred in the diabetic animals at about 60 hr. Estradiol levels were 
higher, but progesterone levels lower, in diabetic than control animals. Administration of 1 
mg of progesterone to diabetic animals 48 hr after PMS resulted in an increase in the number 
of animals ovulating and the number of ova shed. The results indicate that hyperglycemia 
induced by streptozotocin is not inconsistent with production of an LH surge or with ovula- 
tion following ovarian stimulation by PMS. However, the lowered production of progester- 
one, which may be a cause or a result of lowered LH output, appears to be a primary factor 
in the reduced ovulatory rate. 



Female infertility is a well-known conse- 
quence of diabetes mellitus regardless of 
whether the syndrome was induced by pan- 
createctomy or by pharmacologic agents 
(1). A consistent finding, and obviously an 
important factor, in the infertility is a re- 
duction in the percentage of animals 
ovulating and the number of eggs ovulated 
per animal (2, 3). In recent studies (4, 5) 
immature female rats, made diabetic with 
alloxan, failed to ovulate subsequent to 
stimulation of the ovary by pregnant mare's 
serum gonadotropin (PMS), even though 
follicular maturation and estrogen produc- 
tion were not different from that in control 
animals. Several lines of evidence indicated 
that the site of the primary lesion was the 
hypothalamic -hypophyseal axis. Specifi- 
cally, the lack of an LH surge, secondary to 
a reduced pituitary response to hypotha- 



' To whom all correspondence should be addressed. 



lamic LH releasing hormone, was inter- 
preted as the diabetes-induced defect (5). 
The present study was undertaken to ex- 
amine the ovarian steroidogenic and ovu- 
latory responses, as well as changes in 
serum levels of LH, in animals made diabet- 
ic with the antibiotic streptozotocin. The 
latter drug is less toxic than alloxan, but 
very effective at removing pancreatic beta 
cell activity (6). 

The results demonstrate that treatment 
with streptozotocin causes a typical syn- 
drome of diabetes mellitus and a reduction 
in the ovulatory LH surge in immature rats 
injected with PMS. The number of animals 
ovulating and the number of ova shed are 
reduced in these diabetic animals but both 
indices of ovulatory function can be re- 
stored to normal by the administration of 
progesterone 48 hr after the PMS. 

Materials and Methods. Immature 
(26-30 day) female rats of the Holtzman 
strain were housed 5-10 animals/cage in 



207 



208 



RESPONSE TO PMS IN THE DIABETIC 



temperature (25 ± 2°) and light (12 hr light/ 
day) controlled quarters and given free ac- 
cess to food and water. Diabetes mellitus 
was induced by the intravenous injection 
(iv) of streptozotocin (Sigma Chemical Co., 
St. Louis, Mo.). The drug was dissolved 
immediately before use in 0.01 M citrate 
buffer (pH 4.5) and used at a dose of 60 
mg/kg body weight; controls received only 
the buffer. The animals which received the 
drug demonstrated polydipsia and poly- 
phagia within 48 hr. Hyperglycemia was 
verified at the time of sacrifice by deter- 
mination of the concentration of serum glu- 
cose using the glucose oxidase method, 
with kits obtained from Sigma. Animals 
having serum glucose concentration levels 
more than twice that of controls were con- 
sidered diabetic. 

Three to four days after administration of 
streptozotocin, 20 lU of PMS (Sigma), dis- 
solved in 0.15 M NaCl, was injected (iv) 
while the animals were under light ether 
anesthesia. The animals were decapitated 
at various times after injection of the PMS 
and the blood from the trunk collected in 12 
X 75-mm glass tubes. After clotting at room 
temperature, the blood was centrifuged at 
2000^ for 20 min and the serum stored at 
-20° until assayed for hormone content. 

The ovaries were removed as quickly as 
possible after decapitation, cleaned of 
adhering tissue and weighed to the nearest 
0.1 mg on a torsion balance. When 17a- 
hydroxylase activity was to be measured 
the ovaries were quickly frozen, and 
stored, at -20°. In animals killed at 72 hr 
after PMS the oviducts were flushed with 
0.15 A/ NaCl containing hyaluronidase into 
a depression slide and the number of ova 
counted with the aid of a dissecting micro- 
scope. 

Serum LH concentrations were deter- 
mined by double antibody radioimmunoas- 
says using kits supplied by NIAMDD-Rat 
pituitary program: details of the procedure 
have been reported previously (7). Anti-rat 
LH antiserum No. 3, with an initial dilution 
of 1:10,000, was used with rat LH-RP-1 as 
the standard. Hormones were iodinated by 
a modification of the Butt method (8). 

Serum levels of estradiol and progester- 



one were determined by radioimmunoas- 
say. Tritiated estradiol, (2,4,6,7-»H(N)es- 
tradiol-17j3; 115 Ci/mmole) and progester- 
one (l,2,6,7-3(N)progesterone; 97.9 Ci/ 
mmole) were purchased from New England 
Nuclear Corporation (Boston, Mass.) and 
used without further purification. Anti- 
estradiol antiserum was obtained from 
Dr. D. Exley (University of Liverpool, 
England) (9) and anti-progesterone anti- 
serum was prepared by Dr. Vernon Ste- 
vens (Ohio State University, Columbus). 
Details of the assay procedure have been 
published (10). 

Ovarian 17a-hydroxylase activity was 
measured by a tritium exchange assay using 
pregnenolone as the substrate (11). The 
ovaries were homogenized in 0. 15 M KCl 
and centrifuged at 10,00Qg for 30 min. The 
supernatant solution was centrifuged at 
105,000!^ for 60 min and the pellet (micro- 
somal fraction) was then resuspended in 
0.15 M Na-K PO4 buffer (pH 7.4). At least 
triplicate samples of this suspension were 
assayed; results were expressed as nano- 
moles pregnenolone converted per hour per 
milligram of microsomal protein. 

Data were subjected to analysis of var- 
iance or x^ analyses. The independent 
Student's / test was applied when appro- 
priate. Differences between means with a 
F value less than 0.05 were considered 
statistically significant. 

Results. In two large groups of rats, the 
60 mg/kg dose of streptozotocin produced a 
mean nonfasting serum glucose concentra- 
tion of 410 ± 27 mg/100 ml {n = 126) com- 
pared to 119 ± 3.2 mg/100 ml {n = 116) in 
the controls. While 24 of the 26 control 
animals killed at 72 hr after injection of 
PMS had ovulated an average of 12.6 ±3.2 
ova, only 6 of the 26 diabetic animals had 
ovulated 11.2 ± 1.2 ova. 

The ovarian weight responses to 20 lU of 
PMS were not significantly different be- 
tween controls and diabetic animals (Fig. 
1). However, the steroidogenic responses 
did shown differences. In Fig. 2 the changes 
in serum estradiol levels are shown for 
groups of 10 rats (experiment A) killed at 
various times after injection of PMS. The 
diabetic animals had the same level of es- 



RESPONSE TO PMS IN THE DIABETIC 



209 



E 

t 50- 



< 

§ 30- 



24 




24 48 72 

TIME(hr) ofttr PMS(20I.U) 

Fig. 1. Ovarian weight increases in control (O) and 
diabetic (x) immature rats of experiments A and B, 
injected (iv) at time with 20 lU of PMS. Diabetes was 
induced by injection (iv) of 60 mg/kg of streptozotocin 
3-4 days before the PMS. Vertical lines represent one 
SEM for groups of 16 rats. 



tradiol 24 hr after PMS but by 48 hr it was 
32% higher than that in controls {P < 0.01). 
The difference in concentration between 
control and diabetic animals was not statis- 
tically significant at 54 hr, but because of 
the rapidly falling values which occurred 
only in controls after 54 hr, differences at 56 
and 58 hr were pronounced. The estradiol 
levels at 60 hr were 95% higher in the 



900- 



1 
S 

M 

Ul 

lOOl 



il 
l| 

h 
II 
II 
1 1 



I! 



10 




DB 



::^ir 



20 30 40 ^ 60 
HOURS AFTER PMS 



70 80 



Fig. 2. The serum estradiol (E2) levels, determined 
by radioimmunoassay, for the control (C) and strep- 
tozotocin (DB)-treated rats of experiment A. All ani- 
mals received 20 I U of PMS at time 0. Vertical lines 
indicate one SEM for groups of 10 rats. 



diabetic animals but by 72 hr they were the 
same in the two kinds of animals. 

The experiment was repeated one month 
later using 6 animals per group (experiment 
B) with quantitatively slightly different re- 
sults, even though the same lot of PMS 
was used in both cases. Ovarian weight in- 
creases were not different from those found 
in animals of experiment A and are included 
with them in the data of Fig. 1. The pattern 
of changes was the same as in rats of ex- 
periment A but higher levels of estradiol 
were found with the animals of experiment 
B (Fig. 3). As with experiment A diabetic 
animals had higher levels of serum estradiol 
than did controls, with the largest differ- 
ences occurring between 54 and 60 hr after 
PMS. While the serum estradiol levels in 
controls in the two experiments were not 
different, the diabetic animals of experi- 
ment B had four times more estradiol than 
did diabetics of experiment A when mea- 
sured at 72 hr. None of the six diabetic ani- 
mals killed at 72 hr had oviductal ova. 



400- 






E300- 


I 


3 

i 


.1 


i 
S 

20O 

M 


m\ 




/y 


\ 
\ 

\ 
\ 

\ 


100- 


1 / 

1 X 


\ 
\ 




/ / 


Sob 




1 / 






1 / 




0. 


»^ *^.c 



10 20 30 40 50 60 
HOURS AFTER PMS 



70 80 



Fig. 3. Estradiol in the sera of females of experi- 
ment B. These animals were injected one month later 
than those of experiment A, with the same dose and lot 
of PMS. Nonfasting serum glucose levels were not 
different, nor was the pattern of change in E2 levels, 
from that of animals in experiment A, but the E2 levels 
were much higher in the females in this experiment. 



210 



RESPONSE TO PMS IN THE DIABETIC 



30 

E 

s 

i 

9 20- 




1000. 



50 60 70 80 
HOURS AFTER PMS 

Fig. 4. Serum progesterone concentrations in the 
animals ( IQ/group) of experiment A. While the pattern 
of change in the diabetic (DB) and control (C) animals 
was similar the latter had significantly more of this 
steroid when measured at 60 and 72 hr after injection 
of PMS. 



Changes in serum progesterone concen- 
trations were obtained for only the animals 
of experiment A (Fig. 4). The trend in the 
changes in concentration of this steroid in 
the two kinds of animals was the opposite 
of that seen for estradiol, i.e., the controls 
had the larger amount. Even after the de- 
crease in serum progesterone which oc- 
curred between 60 and 72 hr in the controls 
the concentrations were more than double 
those of diabetic animals. 

Changes in serum levels of LH for ani- 
mals of experiments A and B are shown in 
Figs. 5 and 6, respectively. At 48 hr after 
PMS serum LH was the same in diabetic 
and control animals: in both the level was 
about four times that of immature controls 
not given PMS. The concentration rose 
quickly to a peak at 58 hr in the control and 
diabetic animals of experiment A but the 
quantitative differences were quite profound 
(Fig. 5). The level of LH in diabetic animals 
killed 72 hr after PMS was nearly three times 
that of controls (P < 0.01). The pattern of 
changes in serum LH in animals of experi- 
ment B were similar to those of experiment 
A but the peak level in B, which was only 
about half that found in animals of experi- 
ment A, was achieved 2 hr later. LH levels 



750 




o-^'A-r 



I ' I ' I 

50 60 70 80 
HOURS AFTER PMS 



Fig. 5. The serum concentration of LH (as RP-1) in 
diabetic (DB) and control (C) rats of experiment A (E2 
levels shown in Fig. 2). 



in diabetic animals of experiment B were 
not significantly different from those found 
in animals of experiment A. 

Previous studies had demonstrated that 
steroid 17a-hydroxylase activity in the 
ovary decreased dramatically following an 
LH surge (11). Assay of the enzyme in im- 
mature diabetic females treated with PMS 



600- 



450- 



300- 




oVa- 



50 60 70 80 
HOURS AFTER PMS 



Fig. 6. Serum LH levels in females of experiment 
B, which had much higher levels of serum E2 (see Fig. 3). 



RESPONSE TO PMS IN THE DIABETIC 



211 



revealed a pattern similar to that found in 
normal females. That is, after an initial de- 
cline in ovarian 17a-hydroxylase activity, 
an increase, which reached a peak at 60 hr 
was seen. The enzyme activity at 72 hr was 
significantly reduced, even though the 
ovaries of the two animals of the group 
which had ovulated were not included in 
the homogenate used for the assay. 

In a third experiment, performed 3 
months later and with a different lot of 
hormone, 7 of 8 control and 6 of 1 1 diabetic 
females had oviductal ova when examined 
at 72 hr after the injection of 20 lU of PMS 
(Table 1). The number of ova shed was re- 
duced by 66% (F < 0.03) in the animals 
treated with streptozotocin. Both kinds of 
females had received (sc) 0. 1 ml of sesame 
seed oil 24 hr before autopsy. Injection of 1 
mg of progesterone, dissolved in 0.1 ml of 
oil, 48 h after administration of PMS in- 
creased slightly (nonsignificantly) the num- 
ber of ova shed by normal females. In the 
diabetic animals, however, ovulation was 
increased by the progesterone to the extent 
that 14 of the 16 ovulated. Furthermore, the 
number of ova released was increased by 
135% (F < 0.05) compared to diabetic an- 
imals not receiving progesterone. With pro- 
gesterone treatment the number of ova was 
not significantly different from the number 
released by normal females. 

Discussion. The results of the present 
study confirm and extend the findings of 
Kirchick et al, (4, 5). These authors found 
that ovarian weight and estrogen produc- 



tion were similar in diabetic and normal 
immature rats when measured about 50 hr 
after a single injection of PMS. Further, 
they noted that serum progesterone levels 
were below normal in diabetic animals 
when measured either at 1200 or 2100 hr on 
the day of presumed proestrus. However, 
in contrast to their results with alloxan 
treatment, the present study has shown that 
animals made diabetic with streptozotocin 
produced an LH surge on the second day 
after injection of PMS and that some of the 
animals ovulated within 72 hr. The lower 
levels of serum LH in the diabetic animals 
may have accounted for the lowered per- 
centage of animals ovulating and the re- 
duced number of ova shed in some diabetic 
animals (Table I). In adult female rats with 
normal estrous cycles only 1 1 to 14% of the 
ovulatory LH surge on proestrus is re- 
quired for ovulation (12). If this is also true 
for immature rats stimulated by PMS we 
would have expected a higher ovulatory 
rate because in both experiments with 
diabetic animals the peak LH levels were 
greater than 30% of those in control ani- 
mals. 

The decrease in ovarian 17a-hydroxylase 
activity seen in diabetic rats between 60 and 
72 hr after PMS (Fig. 7) is another indica- 
tion of an increase in LH release. This de- 
crease in enzyme activity occurs in normal 
females between 48 and 60 hr after PMS, 
immediately following the LH surge. The 
enzyme activity continues to increase in 
hypophysectomized animals until at least 



TABLE I. The Effect of Progesterone upon Ovulation in Normal 
AND Diabetic Immature Rats Injected with PMS 







No. of 


Body 


Serum 


Ova per 


Ovulation 


Treatment 


Group 


rats 


weight 


glucose 


animal 


rate (%) 


Control 


1 


8 


77.5 ±2.1" 


86 ±8" 


14.8 ±4.4" 


87.5 


Diabetic 


2 


11 


62.9 ± 3.0* 


361 ±21* 


4.9 ± 2.4* 


54.5 


Control + 














progest. 


3 


14 


80.4 ± 3.2" 


93 ± 2" 


21.5 ±2.9" 


100 


Diabetic + 














progest. 


4 


16 


57.4 ± 2.3* 


349 ± 2* 


1 1.5 ±3.4" 


87.5 



Note. Results are means ± SEM. Diabetic animals were injected (iv) with streptozotocin (60 mg/kg) 3 days 
prior to injection (iv) of 20 lU of PMS. Progesterone (progest.) was injected (I mg, sc) 48 hr after the PMS. Ova 
were flushed from the oviducts 72 hr after injection of PMS. Groups with the same superscript are not sig- 
nificanUy different from each other, P < 0.05. 



212 



RESPONSE TO PMS IN THE DIABETIC 




-r— r 

12 24 36 46 60 72 

TIME (h) 

Fig. 7. Steroid 17a-hydroxylase activity in the mi- 
crosomal fraction of ovaries from immature female 
rats made diabetic with streptozotocin; I7apH]preg- 
nenolone was the substrate. The pattern of change is 
similar to that for normal females (11) except that the 
decline associated with the LH surge is less pro- 
nounced and is delayed until 72 hr. Controls injected 
with saline (+) had no significant change in enzyme 
level. SEM for triplicate samples does not exceed the 
area covered by the symbols. 



60 hr, and usually to 72 hr after injection of 
PMS before it gradually declines ((11) and 
unpublished data). In the present study, the 
ovarian hydroxylase activity of diabetic 
animals of experiment B at 48 hr post-PMS 
was 36.0 ± 0.4 nmole/mg protein/hr, while 
control ovaries had 32.0 ± 0.4. By 60 hr 
enzyme activity had decreased to 4.7 in 
controls but it had increased to 41.9 in the 
diabetic females. This delay in the ovarian 
response to LH may reflect the lower levels 
of this gonadotropin in the diabetic, and it 
may account for the high serum estrogen 
concentrations seen at 60 hr. However, it 
may also reflect an altered ovarian response 
to the gonadotropin because there is no 
difference in the timing of the LH surge 
between diabetic and normal females 
(Figs. 4, 5). 

Kirchick et al. (5) concluded that the lack 
of an LH surge in alloxan diabetic rats was 
due to a pituitary insensitivity to hypotha- 
lamic gonadotropin-releasing hormone 
(GnRH). Furthermore, they demonstrated 
that the insensitivity was not secondary 
to a lack of estrogen stimulation of the 



pituitary. The present results also indicate 
that the ovaries of the diabetic respond 
to PMS by increasing production of es- 
tradiol. Possibly the estradiol levels were 
too high (Figs. 2 and 3) and rather than 
being stimulatory for LH release they may 
have become inhibitory, as shown by Wyss 
and Pincus (13). 

The stimulatory effect of progesterone on 
ovulation in the diabetic rat suggests that 
the lack of this steroid may be a primary 
factor in the reduction of the LH surge. 
Graybum and Brown-Grant (14) reported 
that a single injection of 0.5 mg of proges- 
terone 50 hr after PMS increased the pro- 
portion of animals ovulating and the 
number of ova shed. The problem is under- 
standing whether the low progesterone 
levels in the diabetic females are a conse- 
quence or a cause for the low LH levels. If 
an ovarian response was altered by the 
diabetic state and progesterone production 
was curtailed we would expect an inhibiting 
efTect on ovulation at the ovarian level (IS). 
On the other hand a reduction in LH release 
may account for the low progesterone 
levels which would result in the lack of an 
increasingly effective positive feedback 
loop. In either situation the administration 
of LH or progesterone would be expected 
to bring about normal ovulation rates; this 
is the case as shown by Kirchick et al, (4) 
and the results in Table I. 

A role for the adrenal in loss of the 
ovulatory response in diabetics must be 
considered. Diabetes has long been recog- 
nized as a stress and to produce elevated 
serum levels of corticosteroids (16). Fur- 
thermore, in immature female rats, similar 
to those used in the present experiments, 
adrenal secretion of corticosterone was in- 
creased in response to PMS administration 
(17). The level was not high enough to in- 
hibit ovulation, however, and after the LH 
surge in these animals adrenal function de- 
creased dramatically: an ovarian product 
was responsible for the latter effect (17). If 
the progesterone from the heavily lutein- 
ized ovaries was responsible for lower- 
ing of the serum corticosterone levels, a 
point which has not been clearly estab- 
lished (see discussion in (17)), then the 



RESPONSE TO PMS IN THE DIABETIC 



213 



Stimulatory effect of progesterone on ovu- 
lation in the diabetic may involve yet an- 
other mechanism in addition to its effect on 
the release of LH and its direct effects upon 
the ovary. We have not measured serum 
corticosterone levels in the immature dia- 
betics used in the present experiments. 

In conclusion, the present study has 
shown that hyperglycemia induced by 
streptozotocin is not inconsistent with an 
LH surge or with ovulation. The difference 
in responses to those found with hyper- 
glycemia induced by alloxan (4, 5) may 
be due to differences in the degree of pan- 
creatic beta cell destruction induced by 
streptozotocin and consequently to the se- 
verity of the diabetes induced. Under- 
standing the causes for ovulation reduction 
in uncontrolled diabetic animals will obvi- 
ously require investigations into ovarian re- 
sponses as well as those involving the 
hypothalamic -hypophyseal axis. 

Thanks arc due to Pierre Kremers of the University 
of Liege for supplying the substrate used in the 17a- 
hydroxylase assay to Drs. D. Exley and V. Stevens for 
anti-steroid antibodies and to the Rat Pituitary Pro- 
gram of the NIAMDD, National Institutes of Health, 
for supplying radioimmunoassay kits. The expert 
technical assistance of Mrs. Murriel Wagoner is 
gratefully acknowledged. 

1. Chiere RA, Pivetta OH, Foglia VG. Altered ovu- 
lation patterns in experimental diabetes. Fertil 
Steril 20:661-666, 1969. 

2. Davis EM, Fugo NW, Lawrence KB. Effect of 
alloxan diabetes on reproduction in the rat. Proc 
Soc Exper Biol Med 66:638-641, 1947. 

3. Lawrence DM, Contopoulos AN. Reproductive 
performance in the alloxan diabetic female rat. 
Acta Endocrinol 33:175-184, 1960. 

4. Kirchick HJ, Keyes PL, Frye BE. Etiology of 
anovulation in the immature diabetic rat treated 
with pregnant mare^s serum gonadotropin: Ab- 
sence of the preovulatory LH surge. Endocrinol- 
ogy 102:1867-1873, 1978. 

5. Kirchick HJ, Keyes PL, Frye BE. An explanation 
for anovulation in immature alloxan-diabetic rats 



treated with pregnant mare*s serum gonadotropin: 
Reduced pituitary response to gonadotropin- 
releasing hormone. Endocrinology 105:1343- 
1349, 1979. 

6. Hoftiezer V, Carpenter A.-M. Comparison of 
streptozotocin and alloxan induced diabetes in the 
rat, including volumetric quantitation of the pan- 
creatic islets. Diabctologia 9:178-184, 1973. 

7. Mallampati RS, Johnson DC. Serum and pituitary 
prolactin, LH and FSH in androgenized female 
and normal male rats treated with various doses of 
estradiol benzoate. Neuroendocrinology 11:46- 
56, 1973. 

8. Johnson DC. Maintenance of functional corpora 
lutea in androgenized female rats treated with 
PMSG. J Reprod Fertil 56:263-269, 1979. 

9. Exley D, Johnson AW, Dean PDG. Antisera 
highly specific for 17/3-estradiol. Steroids 
18:605-620, 1971. 

10. Grotjan HE Jr, Johnson DC. Temporal variations 
in reproductive hormones in the immature male 
rat. Proc Soc Exp Biol Med 152:381-384, 1976. 

11. Johnson DC. Temporal changes in ovarian steroid 
17a-hydroxylase in immature rats treated with 
pregnant mare's serum gonadotropin. Proc Soc 
Exper Biol Med 159:484-487, 1978. 

12. Grieg F, Weisz J. Preovulatory levels of luteiniz- 
ing hormone, the critical period and ovulation in 
rats. J Endocrinol 57:235-245, 1973. 

13. Wyss HT, Pincus G. Effect of pregnant mare's 
serum gonadotropin, estradiol and progesterone 
on superovulation in the immature rat. Endocri- 
nology 75:586-591, 1964. 

14. Graybum JA, Brown-Grant K. The role of oestro- 
gen in the induction of ovulation in the 
gonadotrophin-treated immature rat. J Endocrinol 
42:409-416, 1968. 

15. Takahashi M, Ford JJ, Yoshinaga K, Greep RO. 
Induction of ovulation in hypophysectomized rats 
by progesterone. Endocrinology 95:1322-1326, 
1974. 

16. Devecerski MC, Frawley TF. Adrenal steroid 
production in rats with alloxan diabetes. Endocri- 
nology 73:368-391, 1963. 

17. Ramaley JA, Olson J. Adrenal function in rats 
given PMS before puberty: Response to ether 
stress. Neuroendocrinology 14:1-13, 1974. 



Received June 17, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 214-220 (1982) 

Carbohydrate Metabolism during the Postprandial Intestinal Hyperemia (41501) 
ROBERT H. GALLAVAN, Jr.,' and C. C. CHOU 

Departments of Physiology and Medicine. Michigan State University, East Lansing, Michigan 48824 



Abstract. Intestinal carbohydrate uptake and utilization were studied before and during 
the perfusion of the canine jejunal lumen with food. Under conditions of normal oxygen 
consumption (1.8 ml Oj/min/lOO g) and blood flow (46.4 ± 3.5 ml/min/100 g), intestinal 
glucose uptake was low (2 mg/min/100 g) and glucose utilization was aerobic. When food was 
present in the lumen, there was a 15% increase in intestinal metabolism, as measured by 
oxygen consumption, while lactic acid production increased threefold. This increase in lactic 
acid production was not due to a shift in glycolysis toward lactic acid synthesis as the venous 
lactic acid/pyruvic acid concentration ratio actually decreased. Intestinal glucose utilization 
was offset by carbohydrate absorption as glucose uptake from arterial blood decreased. The 
data suggest that a relatively greater proportion of intestinal energy demand is met by 
glycolysis during the absorption of nutrients than at rest and supports the findings of other 
investigators that a portion of absorbed glucose is metabolized. In addition, the study indi- 
cates that intestinal glucose metabolism does not terminate in lactic acid synthesis as 
suggested by a previous study. 



Recent studies indicate that the postpran- 
dial intestinal hyperemia is limited to those 
portions of the small intestine which have 
been exposed to chyme (1, 2). It has also 
been shown that the constituents of chyme 
which are responsible for this hyperemia 
are the products of enzymatic food diges- 
tion (3). Of these, glucose has been shown 
to increase jejunal blood flow and oxygen 
consumption when placed in the lumen in 
solution (4, 5). A number of studies have 
indicated that a portion of the glucose 
which is absorbed by the small intestine is 
metabolized (6-9) and Sit et al. (4) have 
shown that this metabolism is responsible 
for a significant portion of the glucose- 
induced increase in jejunal blood flow and 
oxygen consumption. 

In two studies of glucose absorption (8, 
9), a small fraction of the absorbed glucose 
appeared as lactic acid in the venous blood; 
however, it is not known if this lactic acid 
represents the end product of intestinal glu- 



' To whom all correspondence should be addressed: 
Departmeni of Physiology (ML 576), College of 
Medicine. University of Cincinnati, Cincinnati, Ohio 

45267. 



cose metabolism, as suggested by Lester 
and Grim (10), or simply the by-product of 
aerobic glycolysis. In addition, in previous 
studies of intestinal glucose absorption, 
glucose was the only nutrient present in the 
lumen and it is not certain to what extent 
absorbed glucose is utilized for intestinal 
energy production when other substrates 
are available. Therefore, in this study we 
examined the nature of carbohydrate utili- 
zation in the canine jejunum when either 
saline or a nutrient solution containing fat, 
carbohydrate, and protein were present in 
the lumen. 

Materials and Methods. Adult mongrel 
dogs of either sex (15-25 kg; N = 12) were 
fasted for 24 hr, anesthetized with sodium 
pentobarbital (30 mg/kg, iv) and ventilated 
with a positive pressure Harvard respirator 
to ensure normal arterial pH (7.38-7.43). 
Following a midline abdominal incision, a 
loop of the jejunum about 30 cm distal to 
the ligament of Treitz was exteriorized and 
a segment drained by a single vein was 
selected. After administration of sodium 
heparin (6 mg/kg), the vein draining the 
segment was cannulated and the venous 
eflluent was directed through an extracor- 
poreal electromagnetic flow transducer 



214 

^7-9727/S2/I()02l4'07$01.()0/0 
,nght .<■■■ 1982 by the Society for Experimental Biology and Medicine. 
y^f^f.* reserved 



JEJUNAL CARBOHYDRATE METABOLISM 



215 



(Biotronex BLC-2048-E04 connected to a 
Biotronex BL-610 flowmeter) to a venous 
reservoir containing 200 ml of a dextran so- 
lution (6% in normal saline). The volume of 
the reservoir was maintained at 200 ml by 
returning the contents to the femoral vein at 
a rate equal to the venous outflow. 

A piece of rubber tubing was placed in 
each end of the jejunal segment and tied in 
place. Both ends of the segment were tied 
and cut and the mesentery was cut to 
exclude collateral flow. The segment was 
covered with a plastic sheet and kept at ST" 
with a heat lamp. The tubing in the proxi- 
mal end of the segment was connected via a 
Masterflex pump (Cole Parmer, Chicago, 
HI.) to a reservoir containing normal saline 
at 37°. Warm saline was perfused through 
the segment at a rate of 6 ml/min. The tub- 
ing in the distal end of the segment served 
as a drain for the segment. 

The arterial -venous oxygen (A-VO2) 
content difTerence across the segment was 
measured continuously by perfusing femo- 
ral arterial blood and a portion of the ve- 
nous outflow from the segment through the 
cuvettes of an A-VO2 content difference 
analyzer (A VOX Systems, San Antonio, 
Tex.). The analyzer was previously cali- 
brated with a Lex-Oz Con TL oxygen con- 
tent analyzer (Lexington Instruments, 
Waltham, Mass.) and the output signal was 
recorded continuously on a Hewlett-Pack- 
ard recorder (Waltham, Mass.). The signal 
from the flowmeter was also recorded con- 
tinuously and the flow transducer was cali- 
brated periodically during the course of the 
experiment by measuring the venous out- 
flow with a graduated cylinder and stop- 
watch. Systemic arterial pressure was 
monitored continuously through a femoral 
artery cannula. 

After surgery had been completed, saline 
was perfused through the lumen until blood 
flow and the A - VO2 content diffierence had 
reached a steady state (approx. 30 min). At 
that time, duplicate blood samples were 
taken from the femoral artery and the ve- 
nous outflow of the segment for the mea- 
surement of blood pH, PO2, pCOa, lactic 
acid, pyruvic acid, and glucose. The perfu- 
sate was then changed to a mixture of di- 
gested food and bile, at 37", and, when a 



new steady state had been achieved, dupli- 
cate arterial and venous blood samples 
were taken as before. The animal was then 
killed with an overdose of anesthetic and 
the segment was excised, trimmed of all 
mesentery, and weighed. Blood flow, oxy- 
gen consumption, glucose uptake, and lac- 
tic acid and pyruvic acid production were 
expressed in units per 100 grams tissue 
weight. 

Care was taken during the collection of 
the arterial and venous blood samples for 
the measurement of pH, pOg, and PCO2 to 
ensure that the samples were not contami- 
nated with room air. Arterial and venous 
blood samples for the determination of arte- 
rial and venous glucose, lactic acid, and 
pyruvic acid concentrations were collected 
on ice in test tubes containing 10 mg of 
NaF. The NaF served to inhibit red blood 
cell glycolysis. The portion of blood to be 
used for the measurement of the pyruvic 
acid concentration was pipetted from each 
blood sample immediately after collection 
and vortexed with an 8% perchloric acid 
solution to precipitate the blood proteins. 
This served to block enzymatic degradation 
of the pyruvic acid. 

The arterial and venous blood gases were 
measured with a BMS 3 MK II Blood Mi- 
croSystem and Acid -Base Analyzer (Lon- 
don Co., Cleveland, Ohio). The glucose 
concentration in each blood sample was 
measured in a YSI Glucose Analyzer (YSl, 
Yellow Springs, Ohio). The pyruvic acid 
and lactic acid concentrations were mea- 
sured spectrophotometrically using a lactic 
acid dehydrogenase assay (Sigma Chemical 
Co., St. Louis, Mo.). 

The oxygen consumption of the segment 
was determined by multiplying the appropri- 
ate values from the blood flow and A - VO2 
content difference recordings after cor- 
recting for the lag time between the re- 
cordings due to the distance between 
monitoring points. Glucose uptake was cal- 
culated as the product of the arterial -ve- 
nous concentration difference and jejunal 
blood flow while lactic acid and pyruvic 
acid production were taken as the product 
of the venous -arterial concentration dif- 
ference and blood flow. 

The nutrient solution used in this study 



216 



JEJUNAL CARBOHYDRATE METABOLISM 



TABLE 1 . Mean Systemic Arterial Pressure (SAP) and Jejunal Blood Flow (BF), A - VO, Content 
Difference (AA VO2), Oxygen Consumption (VO,), and Venous Hematocrit (Hct) before 

AND DURING THE PERFUSION OF FOOD THROUGH THE LUMEN 









Lumen contents 








Saline 


Food 


Food-saline 


SAP (mm Hg) 

BF (ml/min/lOO g) 

AAVO,(mlO,/100ml) 

VO2(mlO,/min/100g) 

Hct 




117 ±5 
46.4 ± 3.5 

4.1 ±0.3 
1.78 ±0.06 

41 ± 1 


117±5 
54.6 ± 4.2 

4.0 ± 0.3 

2.06 ±0.10 

40±2 


0±3 
8.2 ± 1.3* 
-0.1 ±0.1 
0.28 ± 0.05* 
-1±2 


Note. Values are means ± 
♦ P < 0.05. 


SEM; N = 


= 12. 







contained equal parts by weight of fat, 
protein, and carbohydrate. Fifty grams of 
this food mixture was combined with SOO 
mg of a pancreatic enzyme preparation 
(Viokase, Viobin Co., Monticello, 111.) and 
mixed at room temperature for 5 hr in 400 
ml of 0. 1 yv NaHCOj. Prior to each experi- 
ment, nine parts of the digested food solu- 
tion were mixed with one part of bile from 
the dog's gallbladder. The osmolarity of the 
final solution was adjusted to 300 ± 20 
mosmole/kg with NaCl or distilled water 
and the pH was adjusted to 7.0 ± 0.2 with 
HCI or NaOH. 

The data are expressed as the mean ± 
SEM and comparisons between treatments 
were made using Student's / test modified 
for paired comparisons. Statistical signifi- 
cance was set at P values less than 0.05. 

Results. As seen in Table I, mean sys- 
temic arterial pressure was 117 ± 5 mm Hg 
and it did not change during the course of 
the experiment. Also, as shown in Table I, 
jejunal blood flow was 46.4 ± 3.5 ml/min/ 
100 g and the A-VO2 content difference 
was 4. 1 ± 0.3 ml Og/lOO ml when saline was 
in the lumen yielding an oxygen consump- 
tion of 1.78 ± 0.08 ml Og/min/lOO g. Glu- 
cose uptake was 1.96 ± 0.24 mg/min/100 g 
and jejunal lactic acid and pyruvic acid pro- 
duction were 0.26 ± 0. 10 mg/min/100 g and 
0.059 ± 0.012 mg/min/100 g, respectively 
(Fig. 1). The venous lactic acid/pyruvic acid 
concentration ratio {L/P ratio) was 26 ± 2 
(Fig. 1). This indicates that when saline is 
present in the lumen, jejunal glucose me- 
tabolism is aerobic and does not terminate 
in lactic acid synthesis (11). 



c » 

00 
»— o 



o ^ c 0- 

00 -2.0 

o e 

-4.0 



2co 
u o o 

<Z ^ 

•ft » 



■o 
— Co 

III 

0- E 



2.0 

1.0 



0.2 

0.1 



20H 

oJ 



40 

20- 

0- 



i? 



1 



rj 



i 



l(^=p<0.0S 



m ' .i] v 



N3 F F-NS 



NS F F-NS 



NS F F-NS 



NS F F-NS 



NS F F-NS 

Fig. 1. Mean ± SEM of jejunal glucose consump- 
tion and lactic acid and pyruvic acid production and 
the arterial and jejunal venous lactic acid/pyruvic acid 
concentration ratios with either normal saline (NS) or 
food (F) in the lumen. 



JEJUNAL CARBOHYDRATE METABOLISM 



217 



TABLE IL Arterial and Venous Blood pH. 

pO„ ANDpCO, BEFORE AND DURING PERFUSION 

OF THE Jejunal Lumen with Food 



Arterial 



Venous A-V Difference 



Saline in the lumen 
pH 7.41 ± 0.01 7.40 ± 0.01 0.01 ± 0.004* 
pO, 86 ± 4 37 ± 1 49 ± 4* 

pCO, 33 ±1 34 ±1 -1±1 

Food in the lumen 

pH 7.41 ± 0.01 7.38 ± 0.01 0.03 ± 0.04*t 

pOt 85 ±4 38 ±2 47 ±3* 

pCO, 34 ± 2 38 ± 2 -4 ± l*t 

Note. Values arc means ± SEM; N = 12. 
♦ P < 0.05. 

f P < 0.05 relative to the corresponding value with 
saline in the lumen. 



The arterial pH was 7.41 ± 0.01, thepOg 
was 86 ± 4 mm Hg and thepCOg was 33 ± 1 
mm Hg (Table II). The venous pH (7.40 ± 
0.01) was only slightly less than arterial pH 
but the difference was statistically signifi- 
cant. In addition, the venous pOz (37 ± 1 
mm Hg) was significantly less than arterial 
PO2 but there was no significant ar- 
terial -venous pCOj difference. The venous 
hematocrit was 41 ± 1 (Table I). 

When food was perfused through the 
lumen, there was a significant increase in 
jejunal blood flow and oxygen consumption 
(Table I). Jejunal blood flow increased 8.2 
± 1.3 ml/min/100 g (18 ± 3%) to 54.6 ± 4.2 
ml/min/100 g. Oxygen consumption in- 
creased 0.28 ± 0.05 ml Oj/min/lOO g to 2.06 
± 0.10 ml Ojj/min/lOO g, an increase of 15 ± 
3%. The increased demand for oxygen was 
met entirely by the increase in blood flow as 
the A-VO2 content difference did not 
change (Table I). 

The luminal presence of food signifi- 
cantly decreased jejunal glucose uptake 
from the blood to -0.88 ± 0.63 mg/min/lOO 
g, a decrease of 2.84 ± 0.68 mg/min/lOO g 
(Fig. 1). At the same time both lactic acid 
and pyruvic acid production increased sig- 
nificantly. Lactic acid production increased 
to 1.02 ± 0.18 mg/min/lOO g, an increase of 
0.76 ± 0.17 mg/min/100 g or nearly 300%. 
Pyruvic acid production increased by 0.07 
± 0.024 mg/min/100 g to 0.128 ± 0.027 



mg/min/100 g. There was no significant 
change in the arterial L/P ratio while the 
venous L/P ratio decreased slightly ( -4 ± 1). 

There were no significant changes in 
arterial pH, pOg, or pCOj when food was 
present in the lumen (Table II); however, 
there were significant increases in the ar- 
terial-venous pH and pCOj differences. 
The arterial -venous pCOj difference in- 
creased from - 1 ± 1 to -4 ± 1 mm Hg and 
the arterial -venous pH difference in- 
creased from 0.01 ± 0.004 to 0.03 ± 0.004, 
reflecting the increased CO2 and lactic acid 
production. There was no significant 
change in either the arterial -venous PO2 
difference (Table II) or the venous hemato- 
crit (Table I). 

Discussion. Recent studies indicate that 
when glucose is placed in the lumen of the 
small intestine, 15-40% of the glucose 
which is absorbed does not appear in the 
venous blood (6, 7, 9). This net loss of glu- 
cose across the intestinal wall has been at- 
tributed to the utilization of absorbed glu- 
cose for energy production within the 
mucosal layer (8- 10). This concept is sup- 
ported by the fact that there is an increase 
in the venous lactic acid concentration 
during glucose absorption which is equiva- 
lent to 5- 10% of the absorbed glucose (8, 
9). It has even been suggested that this in- 
tracellular conversion of glucose to lactic 
acid in the small intestine and its sub- 
sequent release into the portal circulation 
where it can be utilized as a substrate for 
hepatic gluconeogenesis may serve as an- 
other mechanism of intestinal glucose ab- 
sorption (12). 

In this study, we examined the nature of 
jejunal glucose metabolism with either 
saline or food in the lumen in order to de- 
termine if glucose utilization increases 
when a variety of nutrients are present in 
the lumen. Furthermore, we wished to de- 
termine if lactic acid is the end product of 
intestinal glucose metabolism as suggested 
by Lester and Grim (10). 

The carbohydrate used in this study was 
sucrose (4 g/100 ml) which consists of one 
molecule each of glucose and fructose and 
is broken down to its constituents by the 
intestinal brush border enzyme, sucrase. 



218 



JEJUNAL CAEBOHYDRATE METABOUSM 



Although fructose was not measured in this 
^tudy« 60-95% of absorbed fructose is 
converted to either glucose or lactic acid 
within the small intestine (9, 13- 15) so that^ 
in this experiment, venous glucose and lac- 
tic acid represent at least 80% of the ab- 
sorbed carbohydrate. 

The data indicate that, when saline is 
present in the lumen, there is a low level of 
glucose uptake and lactic acid and pyruvic 
acid production. These findings support 
those of Shoemaker et al. (9), who reported 
a control lactic acid production of only 2.8 
mg/min for the entire nonhepatic splanchnic 
area in conscious dogs, and contradict 
those of Lester and Grim ( 10), who found 
that in canine jejunal mucosa 80% of me- 
tabolized glucose was converted to lactic 
acid in vitro. The reasons for this discrep- 
ancy in results between in vivo and in vitro 
studies are not readily apparent. It is possi- 
ble that portions of the tissues in the in vitro 
studies were anoxic and relied on anaerobic 
glycolysis for energy production; however, 
Wilson (12) presents strong evidence 
against such a possibility in similar studies 
using everted rat gut sacs. 

It is interesting to note that Wilson has 
also reported a sixfold increase in lactic 
acid production by everted rat ileal sacs 
when glucose was placed on the mucosal 
rather than the serosal side (12). Therefore, 
it is possible that the difference between the 
in vivo andm vitro studies of canine jejunal 
glucose metabolism may be due to the fact 
that in the in vitro studies glucose was ex- 
posed to the mucosal surface while in the 
control period of the in vivo studies it was 
not. It is not clear, however, if these results 
arc applicable to other species as the nature 
of glucose metabolism in the rat and mouse 
intestines is considerably different from 
other species such as the guinea pig, ham- 
ster, and rabbit (12). 

When food was perfused through the 
lumen, there was an increase in jejunal 
oxygen consumption, a decrease in glucose 
uptake from arterial blood, and an increase 
in both lactic acid and pyruvic acid produc- 
tion. As it was not possible to accurately 
measure the disappearance of carbohydrate 
from the lumen in these studies, the conclu- 



sions which may be drawn from the avail- 
abie data must be based oo certain assump- 
tions and are therefore tentative. 

In the saline-perfiised jejunal loop, ^u- 
cose was taken up firom the blood at a rate 
of 66 /imole Cmm-'- 100 g-'. Of this, lactic 
acid and pyruvic acid production accounted 
for II /imole Ojmin"'100 g"*. Oxygen 
consumption during this period was 80 
/imole O2'min~'-100 g~'. If the remaining 
55 /imole C-"'-100 g"' were converted 
entirely to COj, then 70% of the oxygen 
consumed by the segment was used for ^u- 
cose metabolism. However, Lester and 
Grim (10) have reported that only 50% of 
jejunal oxygen consumption is us^ for car- 
bohydrate metabolism in vitro. If these re- 
sults are representative of conditions in 
vivo, then approximately 25% of the glu- 
cose metabolized by the saline-perfiised 
jejunal loop was converted to products 
other than CO2, lactic acid, and pyruvic 
acid. 

When food was perfused through the 
lumen of the jejunal loop, there was a 
threefold increase in the rate of lactic acid 
production by the segment. This increase 
could have been due to either a shift in car- 
bohydrate metabolism toward lactic acid 
synthesis while total carbohydrate metabo- 
lism remained unchanged or to an increase 
in the overall rate of carbohydrate metabo- 
lism. The former conclusion would support 
Wilson's findings that the metabolic fate of 
glucose depends upon which surface of the 
cell it must cross, i.e., serosal or mucosal. 
However, a shift in carbohydrate metabo- 
lism toward lactic acid synthesis is usuaUy 
associated with an increase in the venous 
LIP ratio (11). This did not occur, rather the 
venous LIP ratio decreased slightly in this 
study. Therefore, it would seem likely that 
the increased rate of lactic acid synthesis 
represents an increase in the overall rate of 
carbohydrate metabolism. 

There is some evidence to support the 
conclusion that a portion of absorbed glu- 
cose is metabolized by the canine jejunum 
in a recent study by Sit et al. (4). In that 
study, the authors examined the effect of 
placing either glucose or S-O-methyl glu- 
cose in the jejunal lumen on jejunal oxygen 



JEJUNAL CARBOHYDRATE METABOLISM 



219 



consumption. While both glucose and 
3-0- methyl glucose are actively transported 
by the mucosal epithelia, only glucose is 
metabolized by intestinal tissues. Sit et aL 
found that although the jejunal loops ab- 
sorbed equal amounts of each carbohy- 
drate, only glucose increased intestinal 
oxygen consumption. The authors attrib- 
uted this difference in response to metabo- 
lism of absorbed glucose. 

If jejunal carbohydrate metabolism did 
increase when food was perfused through 
the lumen, the amount of absorbed carbo- 
hydrate which was metabolized depends 
upon which assumptions regarding the na- 
ture of intestinal carbohydrate metabolism 
which we have discussed previously are 
correct. If, as Lester and Grim have re- 
ported (10), 50% of intestinal oxygen con- 
sumption is directed toward carbohydrate 
metabolism, then total carbohydrate utili- 
zation in the food-perfused loops was 
84 Mmole Cmin'^lOO g* (38 ^mole 
Cmin"*100 g"' converted to lactate/ 
pyruvate; 46 ^imole Cmin"^100 g"* con- 
verted to CO2). This would represent an in- 
crease in the rate of carbohydrate metabo- 
lism over control of 18 ^imole Cmin~*- 100 
g~* or 16% of the absorbed glucose. The 
rate of glucose absorption in this case being 
equal to the sum of the decrease in the rate 
of glucose uptake from the blood (96 ^imole 
C-min"^' 100 g~0 and the assumed increase 
in the rate of carbohydrate metabolism (18 
/Ltmole Cmin-*- lOOg-^. If 70% of the oxy- 
gen consumed were directed toward carbo- 
hydrate metabolism or if glucose were de- 
graded to compounds other than lactic acid 
or pyruvic acid, then the percentage me- 
tabolized would be even greater. 

Although resting glucose utilization by 
the small intestine is low (4,9, 10), it should 
not seem surprising that the rate of gly- 
colysis should increase during periods of 
nutrient absorption. There is no reason to 
believe that absorbed nutrients are com- 
partmentalized as they pass through the 
cells of the mucosa and therefore one would 
expect that a portion of those nutrients 
would be metabolized according to the laws 
of mass action. If the increase in the rate of 
lactic acid production is proportional to the 



overall increase in the rate of glycolysis, 
then it would appear that the relative pro- 
portion of energy demand met by glycolysis 
increased when food was present in the 
lumen as there was a threefold increase in 
the rate of lactic acid production while total 
intestinal metabolism increased only 15%. 
Again, this should not seem surprising as 
glucose is available for immediate utiliza- 
tion while the fats, tripeptides, dipeptides, 
and amino acids require additional catab- 
olism before they can be utilized for energy 
production. 

The results of this study indicate that 
when blood flow and oxygen delivery are 
adequate, there is a low level of intestinal 
glucose metabolism which does not appear 
to terminate in lactic acid synthesis. When 
food is present in the lumen, there is an 
apparent increase in the rate of glucose 
utilization which is proportionally greater 
than the overall increase in the rate of in- 
testinal metabolism. The demand for glu- 
cose by the small intestine is apparently 
offset by carbohydrate absorption as glu- 
cose uptake from the arterial blood actually 
decreases. These findings lend support to 
the conclusions of other investigators (4, 
6-9) that a portion of the glucose absorbed 
from the lumen of the small intestine is 
metabolized and suggests that metabolism 
is directed through the Kreb's cycle rather 
than terminating in lactic acid synthesis. 



1. Gallavan RH Jr, Chou CC, Kvietys PR, Sit SP. 
Regional blood flow during digestion in the con- 
scious dog. Amer J Physiol 238:H220-H225, 
1980. 

2. Chou CC, Hsieh CP, Yu YM, Kvietys P, Yu LC, 
Pittman R, Dabney JM. Localization of mesen- 
teric hyperemia during digestion in dogs. Amer J 
Physiol 230:583-589, 1976. 

3. Chou CC, Kvietys P. Post J, Sit SP. Constituents 
of chyme responsible for postprandial intestinal 
hyperemia. Amer J Physiol 235:H677-H682, 
1978. 

4. Sit SP, Nyhof R, Gallavan R, Chou CC. Mecha- 
nisms of glucose-induced hyperemia in the 
jejunum. Proc Soc Exp Biol Med 163:273-277. 
1980. 

5. Shepherd AP. Intestinal capillary blood flow dur- 
ing metabolic hyperemia. Amer J Physiol 
237:E548-E554, 1979. 



220 



JEJUNAL CARBOHYDRATE METABOLISM 



6. p2tnon% DS. Pnctuud JS. A prcpanttion ai per- 
fumed small ifiteMifie for the study atabwrptkm in 
amphibia. J Physiol l^t:405-434. 196g. 

7. Balent GA. Kiss ZF. Varkonyi T. Winmann T. 
Varro. V. Elfecis of prostagbuidin E, and F^ on 
the absorption and portal transport of sugar and 
on the local intestinal ciixuiladon. Prostaglandins 
111:265 268. 1979 

H. Wilson TH. Wiseman G. The use of sacs of 
everted small intestine for the study of the trans' 
ference of substances from the mucosal to the 
serosal surface. J Physiol 123:116- 12S. 1954. 

9. Shoemaker WC. Yanof HM. Tusk LN, Wilson 
TN. Glucose and fructose absorpticm in the un- 
anesthetized dog. Gastroenterology 44c654-663. 
1962. 
10. I>ester RG. Grim £. Substrate utilization and oxy- 
gen consumption by canine jejunal mucosa in 
vitro. Amer J Physiol 229:139- 143. 1975. 



n. Wiliiaflnon JR. Caycoiytic < 

Biol Oiem 244dQZ6-5ll36. 19tt. 

12. Wilson TH. The rale of lactic acid producfioa in 
glucose absorption from the intestine. J Biol 
Oiem 222:751-763, 1956. 

13. BoilmanJL, Mann FC. The utiiizadon of fructose 
following complete removal of the liver. Amer J 
Physiol 96:683-695. I99I. 

14. Ockerman PA. Lunborg H- Conversion of 
fructose to glucose by huimni jcgunum absence of 
galactose-to-giucose converskra. Biochim Bio- 
phys Acta 105J4-.C 1956. 

15. KiyasuJY.ChaikofflL. On the manner of trans- 
port of absorbed fructose. J Biol Chem 224: 
935-946. 1957. 



Received February 9. 1982. P.S.E.B.M. 1982. Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 221-231 (1982) 



Golgi Complex Function in the Excretion of Renal Kallikrein (41502) 

ENRIQUE BRANDAN,' MIREYA ROJAS, NORA LOYARTE, AND 
FERNANDO ZAMBRANO* 

Departamento de Biologia, Facuhad de Ciencias Basicas v Farmac^uticas , Universidad de Chile ^ 

Casilla 653, Santiago, Chile 



Abstract. A Golgi complex rich-fraction containing both N-acetylglucosamine galac- 
tosyltransferase and kallikrein activity has been isolated from kidney of rats previously 
treated with colchicine, a secretion inhibitor, followed by the administration of high-sodium 
solutions, to stimulate biosynthesis or activation of renal kallikrein. After the treatment, 
N-acetylglucosamine galactosyltransferase and kallikrein activities were increased in the 
Golgi complex, about 18- and 24-fold, respectively, as compared to the homogenate. Low 
kallikrein activity was found in the crude light mitochondrial fraction from treated animals, 
whereas a high level of activity was observed in the microsomal fraction. The inverse 
situation was found in rats treated only with high-sodium solution. Results suggest that 
kallikrein is probably transported by microsomal elements, particularly by the Golgi com- 
plex. Furthermore, the evidence seems to indicate that the kallikrein activity reported in the 
plasma membrane and/or in the lysosomal fraction is due to kallikrein secretion, in the form 
of intact granules, which have sedimented with these two fractions. 



Glandular kallikreins are a group of related 
kinin-forming enzymes present at least in 
the major exocrine glands and in the kid- 
ney. Previous studies on the role of the kal- 
likrein -kinin system in the kidney identify 
kinins as potent vasodilators which cause 
natriuresis and diuresis when injected into 
the renal artery (1). 

It is known that urinary kallikrein is a 
glycoproteic enzyme and that the renal kal- 
likrein is secreted into the urine at the level 
of the distal tubule (2, 3). Moreover, both 
the renal and urinary enzymes are im- 
munologically and electrophoretically sim- 
ilar (4). Renal kallikrein is also a glycopro- 
tein and it is found in cells in the form of 
dense granules, which are transported to 
the extracellular space by some undefmed 
mechanism by microsomal elements (5). 

The localization of renal kallikrein is 
subject of dispute. Some authors refer it to 
the lysosomal fraction (6, 7), others to the 



' Present address: Departamento de Biolog(a Celu- 
lar, Universidad Catdlica de Chile, Casilla 114-D, 
Santiago, Chile. 

' To whom all correspondence should be addressed. 



microsomal fraction (8, 9), and still others 
to the plasma membrane fraction of kidney 
homogenates (10). In an attempt to clarify 
the apparently conflicting findings, we have 
isolated and characterized different sub- 
cellular fractions from rat kidney, using 
both differential and density gradient cen- 
trifugation techniques. Our approach was 
to increase the intracellular amount of kal- 
likrein by a combination of two treatments: 
salt loading, that we showed increases the 
activation or the biosynthesis of this en- 
zyme (11), and colchicine treatment, that 
inhibits the liver secretion by a reduction of 
the microtubule content of the cells (12, 13). 

Our results indicate that Golgi complex 
participates in the intracellular transport 
and probably in the biosynthesis and/or ac- 
tivation of renal kallikrein. 

Material and Methods. Female Sprague- 
Dawley rats, 200 to 300 g, fed ad libitum 
were used. Animals were treated with col- 
chicine and then loaded with a 0.342 M 
saline solution at 5% body weight by 
gavage (14). The unanaesthetized animals 
were decapitated 2, 4, 8, and 10 min after 
treatment, then exsanguinated, and their 
kidneys quickly removed, decapsulated. 



221 



0037-9721 (%ll VO^Tl\-\\%^\ .V»\^ 

Ml nghls Te%ervcd. 



222 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



and collected in ice-cold 0.25 M sucrose. 
Colchicine was given ip by way of two 
consecutive injections, 60 and 15 min be- 
fore killing the animals. Two doses of 
colchicine were used in two groups of rats: 
0.5 and 1.0 mg per 100 g body weight, re- 
spectively. UDP-Gal (Calbiochem, Los 
Angeles, Calif.) and UDP-Gal uniformly 
labeled with ^^C in the sugar moiety (New 
England Nuclear Corp., Boston, Mass.) 
were used. The latter was diluted with car- 
rier to a specific activity of about 1 mCi/ 
nmole. 

Cell fractions were prepared from a pool 
of kidneys blotted, weighed, and minced 
with scissors. Centrifugations were carried 
out in Beckmann ultracentrifuge, and all 
operations were carried out at 0-4°. The 
Golgi complex fraction was prepared by the 
method previously described (15), with 
some modifications. About 10 g of minced 
tissue was suspended in 2 vol of 52% su- 
crose containing 0.1 A/ sodium phosphate, 
pH 7.1, and homogenized with three full 
strokes at 1000 rpm using a 50-ml glass 
Potter- El vehjem type homogenizer with 
an i.d. of exactly 1 in and a Teflon pestle 
machined to a diameter of 0.974 in. The 
homogenization was repeated using a pestle 
with a diameter of 0.982 in. The homoge- 
nate was filtered through four layers of 
cheesecloth and adjusted to 43.7% sucrose 
with homogenizing medium. Seven to nine 
milliliters of homogenate were placed in a 
tube and overlaid with sufficient 37.8% su- 
crose to bring the total volume to 15.4 ml. 
This was then successively overlaid with 
6.3 ml of 36% sucrose then with 6.3 ml of 
33% sucrose and finally with 7 ml 29% su- 
crose. The step gradient was then cenlri- 
fuged for 60 min at 25,000 rpm in a SW 25. 1 
rotor. The Golgi complex rich-fractions 1 
and 2 were obtained from the material ap- 
pearing at the 29/33% and 33/36% sucrose 
interfaces, respectively. The fractions were 
diluted with 1/2 vol of cold distilled water 
and the membranes recovered by centrifu- 
gation at 30,000 rpm for I hr in a 30 rotor. 

The light crude mitochondrial, the light 
mitochondrial, and the microsome fractions 
were prepared by the methods of Stein et 
al. (16) as modified by Fleischer and Ker- 



vina (17). About 10 g of minced tissue was 
suspended in 4 vol of 0.25 M sucrose solu- 
tion containing 0.01 M Hepes, pH 7.4, and 
homogenized with three full strokes at 1000 
rpm using a 50-ml Potter- Elvehjem type 
homogenizer (i.d. of glass vessel, 1 in) and 
a Teflon pestle machine to a diameter of 
0.974 in and followed with three full 
strokes with a pestle machined to 0.988 in. 
The homogenate was filtered through a 
110-mesh nylon monofilament bolting 
cloth. The homogenate was differentially 
centrifuged, and a low spin pellet (lOOQg for 
10 min), a light crude mitochondrial, a mi- 
crosomal, and a soluble fraction were suc- 
cessively isolated. The light mitochondria] 
fraction was obtained by further purifica- 
tion of the crude fraction resuspended in 
0.25 M sucrose-0.01 M Hepes-0.001 M 
EDTA, pH 7.4, and recentrifuged for 10 
min at 18,000 rpm in a 40 rotor. The upper 
light portion (mostly heavy microsomes) of 
the pellet was separated and discarded. The 
lower brown portion, enriched in mito- 
chondria, was resuspended in the same so- 
lution and centrifuged again for 10 min at 
14,000 rpm with the same rotor. The re- 
sidual upper layer was again removed. A 
lysosomal fraction was obtained from the 
light crude mitochondrial fraction by step 
sucrose gradient centrifugation according 
to the method of Maunsbach (18). The frac- 
tion, resuspended in 0.3 M sucrose-0.001 
M EDTA, pH 7.0, was layered on a linear 
sucrose gradient (0.3-2.1 M) containing 
0.001 M EDTA and was centrifuged for 180 
min at 24,500 rpm in a SW 25.1 rotor. The 
most dense band is enriched in lysosomes. 

Microsomes were further fractionated 
into a smooth and a rough fraction by a 
modification (17) of the method of Dallner 
(19) as described for liver. Microsomes, re- 
suspended with 0.25 M sucrose-0.015 M 
CsCl solution, were layered into tubes 
containing 1.3 M sucrose-0.015 M CsCl 
and centrifuged for ISO min at 49,000 rpm in 
a 50 rotor. The rough microsomes sediment 
as a pellet and the smooth microsomes re- 
main at the interface. 

Glucose-6-phosphatase was determined 
according to the method of Swanson (20), 
except that incubations were carried out for 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



223 



5 and 10 min. Succinate -cytochrome c re- 
ductase activity was determined as previ- 
ously described (21). Acid phosphatase 
activity, using /3-glycerophosphate as 
substrate, was measured by the method of 
Besseys et aL (22), except that inorganic 
phosphate was measured using the method 
of Chen et aL (23). Galactosyltransferase 
was determined as previously described by 
Fleischer (24). 

Kallikrein activity was measured using 
two different methods: 

(a) By its stimulating effect on uterine 
contractility (25), using bradykinin as a 
standard. Activity is expressed as kallikrein 
equivalent to nanograms of bradykinin per 
milligram of protein. Trasylol (Aprotinin), 
was used as inhibitor of the stimulating ef- 
fect of the enzyme in the bioassay (25). 

(b) By the esterase activity shown by 
kallikrein, using benzoyl-L-arginine ethyl 
ester (BAEE) as substrate (6). The col- 
orimetric reaction was measured by the 
method of Brown (26) and expressed as 
micromoles BAEE hydrolyzed per minute 
per milligram of protein. 

Proteins were determined by Lowry's 
procedure (27), using crystalline bovine 
serum albumin as a standard. Phosphorus 
was determined by the method of Chen et 
aL (23). 

Undiluted aliquots of the sucrose in- 
terphase containing the Golgi complex were 
fixed with 1/lOth of 25% glutaraldehyde 
made up in 0.2 M sodium cacodylate at pH 
7.4, immediately after isolation from the 
step gradient. The other cell fractions were 
fixed by treating an aliquot with an equal 



volume of 5% glutaraldehyde in 0.25 M su- 
crose and 0.2 M sodium cacodylate, pH 7.4. 
After standing overnight in the refrigerator, 
the samples were centrifuged at 20,000 rpm 
for 15 min in a 40 rotor and the supernatant 
discarded. Pellets were then washed twice 
by suspension in 0.25 M sucrose and re- 
centrifuged. Finally the pellets were fixed 
with 1% osmium tetroxide, dehydrated, 
embedded, and sectioned as previously de- 
scribed (28). 

Results. The activity of kallikrein in kid- 
ney homogenate, in light crude mitochon- 
drial, and in microsomal fractions isolated 
by differential centrifugation, is shown in 
Table I. The kidney homogenate of loaded 
rats shows the highest kallikrein activity 8 
min after salt loading. This activity is five 
times higher than that obtained in non- 
loaded rats. Ten minutes after salt loading, 
the kallikrein activity shows only a twofold 
increase with respect to that of nonloaded 
rats. The activity measured 4 min after salt 
loading in light crude mitochondrial and in 
microsomal fractions, isolated from the 
same homogenate, increased 17- and 26- 
fold, respectively, as compared to the val- 
ues obtained in the same fractions from 
nonloaded rats. After 8 min of salt loading 
the corresponding values were increased 
56- and 42-fold, respectively. It is necessary 
to indicate that kallikrein activity in ho- 
mogenate as well as in light crude mito- 
chondrial and in microsomal fractions, from 
kidneys of rats sacrificed 2, 4, 8, and 10 min 
after sham gavage submission, were similar 
to that of nonloaded rats. 

Table II describes the effect of a pre- 



TABLE I. Renal Kallikrein Activity of Light Crude Mitochondrial and Microsomal Fractions 

Obtained from Sodium-Loaded Rats 



Fractions 



Homogenate 



Mitochondrial 



Microsomal 



Nonloaded rat 
2 min after treatment 
4 min after treatment 
8 min after treatment 

10 min after treatment 



1.07 ±0.10 
2.87 ± 0.29 
4.52 ± 0.41 
5.39 ± 0.52 
1.95 ±0.20 



0.75 ± 0.08 

1.97 ±0.16 

12.63 ± 1.04 

42.30 ± 2.97 

21.32 ± 1.94 



0.23 ± 0.03 
0.72 ± 0.08 
6.11 ±0.07 
9.80 ± 1.06 
7.41 ± 0.79 



Note. Results are mean values ± SD of three experiments. Kallikrein activity is in each case expressed as 
kallikrein equivalent to ng of bradykinin/mg of protein, determined by bioassay in cell fractions obtained from 
half of pooled kidneys of 10 loaded rats. 



224 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



TABLE IL Distribution of Renal Kallikrein Activity in Kidney Subcellular Fractions 
Obtained from Sodium-Loaded Rats 





Salt loaded 




Salt loaded and colchicine 






0.5 mg/100 g body wt 


1.0 mg/100 g 


body wt 




Total protein 


Enzyme 


Total protein 


Enzyme 


Total protein 


Enzyme 




(mg) 


activity 


(mg) 


activity 


(mg) 


activity 


Homogenate 


763.8 


4.52 


614.6 


4.31 


694.1 


4.47 


1000^ 














pellet 


339.4 


5.69 


286.2 


5.88 


293.4 


5.68 


Light crude 














mitochondrial 














fraction 


78.9 


12.63 


60.2 


3.61 


79.9 


1.92 


Lysosomes 


10.1 


0.14 


5.8 


0.15 


7.6 


0.15 


Microsomes 


61.1 


6.11 


60.5 


10.11 


59.2 


18.72 


Supernatant 


248.3 


0.28 


211.9 


0.01 


231.6 


0.05 



Note. Values are in each case the average of three experiments obtained in cell fractions from half the kidney 
pool of 10 salt-loaded rats, after 4 min of treatment. Kallikrein activity, measured by bioassay, is expressed as in 
Table 1. Supernatant is the fraction of the homogenate which does not sediment at 100,000^ after 1 hr. 



treatment with two doses of colchicine and 
salt loading for 4 min, in the distribution of 
kallikrein activity in different fractions. The 
highest activities were found in the micro- 
somal fractions which increased nearly 
two- (0.5 mg colchicine) and threefold (1.0 
mg colchicine). The inverse situation is ob- 
served in rats not treated with colchicine, 
which show a higher activity in the light 
crude mitochondrial fraction than in the mi- 
crosomal fraction. Since the composition of 
a subcellular fraction depends on the 
method of isolation, it should be indicated 
that the light crude mitochondrial fraction 
as well as the microsomal fraction, isolated 
by differential centrifugation, are mixed 
fractions. Thus, in addition to mitochon- 
dria, the light crude mitochondrial fraction 
includes lysosomes, granules, and heavy 
microsomes. On the other hand, the micro- 
somal fraction is mostly composed of frag- 
mented rough and smooth endoplasmic re- 
ticulum, including membranes of the Golgi 
complex. As shown in Table II, in sodi- 
um-loaded rats, the pretreatment with 
colchicine induces an increase of activity in 
the microsomal fraction and a decrease in 
the light crude mitochondrial fraction. In 
spite of the different kallikrein content, the 
protein distribution in the different frac- 
tions is only slightly affected, even in rats 



treated with a high dose of colchicine. 
Furthermore, the low kallikrein activity in 
lysosomes, isolated from the light crude 
mitochondrial fraction, remains unaffected 
by the colchicine treatment (see Fig. 3). 

Figures 1 and 2 illustrate the morphology 
of the microsomal and the light crude mito- 
chondrial fractions obtained from sodi- 
um-loaded rats treated and not treated 
with 1.0 mg of colchicine, sacrificed 4 min 
after treatment with salt. A similar and 
typical morphology in both microsomal 
fractions is observed, consisting essentially 
of closed and empty vesicles, ruptured 
vesicles, membrane fragments, and vesicles 
marked by ribosomes attached to their 
membrane. On the other hand, the differ- 
ence in morphology between both mito- 
chondrial fractions is quite clear. Although 
both are rich in mitochondria, the fraction 
obtained from kidneys of rats not treated 
with colchicine shows a larger number of 
dense bodies than the one obtained from 
colchicine-treated rats. Since similar activ- 
ity of acid phosphatase was found in both 
fractions, we suggest that the difference in 
morphology observed is due to the pres- 
ence of granules of secretion or secretory 
vesicles. Thus, the light crude mitochon- 
drial fraction of colchicine-nontreated rats 
seems to be richer in granules or vesicles 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



225 







Fig. I. Electron micrograph of u microsomal fraction obtained from sodium-loaded ral kidney 
homogenate. (a) Untreated with colchicine x 16.480: (b) treated with colchicine, x 16.480. 

Fig. 2. Electron micrograph of a light crude mitochondrial fraction obtained from kidney homoge- 
nate of sodium-loaded rat. (a) Untreated with colchicine, x 1 7,600: (b) treated with colchicine x 17.600. 



than the mitochondrial fraction of col- nents, obtained from pooled kidneys of 10 

chicine-treated rats. rats, treated with I.O mg of colchicine, and 

Table III illustrates the kallikrein activity killed 4 min after sodium loading. Since 

of the microsomal fraction and its compo- most kallikrein activity resides in micro- 



'JJ. 



'.'.vr.v ■?.». '"i i>__.i*i." £x:»i">:'v 



rlTirFi 






b/ 






4 



i.^-:. 



'^> 







I If, i l.h-' riori MiKrii^:r;iph of :i lysDsomal fraction nhtaincJ from the light crude mitochondrial 
fi.i' lion lit I ly ? f.ii t.iiircaiciJ wiih colchitinc. x |H.4(Mr <hi treated uiih colchicine. > 18.400. 

I I'. I III • Iron fiii(.ro^'r.ipli ot :i (ttt\iii n)mplex-nch fi action (equivalent to fraction I -^ 2) obtained 
»., .r« f. IP ii».r {'i,i(iiriii of MKliurii hiadeil lai kidnev homoiicnate. lai I ntreated with colchicine, 

I', K'Mi M.) lr«-.il'i} .'.ilh .1.1' III- iiK-. • \t,.MH). 

Mini- .f .It hihic 11). ihc cii/ynic was mca- nc\s. Kallikrein activii\ was also measured 

MiM'<| in lotJii Miicii>s<iiiK's as well as in mi- in the (io\\i\ complex-rich fractions 1 and 2, 

< in-.oiiK- . flivKJcd inin smooth and lough obtained from the remaining half of the 

fiai lion-.. (»l»i:iiind hnin [»all the p«)()led kid- pooled kidneys. Ihe kallikrein activity in 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



227 



TABLE in. Renal Kallikrein Activity of Different Components of the Microsomal Fraction of 

Salt-Loaded Colchicine-Treated Rats 





Colchicine 








(l.Omg/lOOgwt) 




Total protein 








(mg) 




Enzyme activity 


672.0 






4.36 


54.45 






18.83 


39.94 






2.90 


10.74 






70.24 


1.74 






134.26 


2.79 






175.35 


220.42 






0.04 



Homogenate 

Microsomes 

Rough microsomes 

Smooth microsomes 

Golgi complex-rich fraction 1 

Golgi complex-rich fraction 2 

Supernatant 



Note. Fractions were isolated from the same pool of kidneys from 10 treated rats. Values are the mean of three 
experiments and are expressed as in Table I. Total microsomes, smooth and rough microsomes, and the super- 
natant were obtained from half the pooled kidneys. The remaining tissue was used to isolate Golgi complex-rich 
fractions. The protein content of each fraction is referred to as the whole homogenate. 



the smooth microsome fraction was almost 
24-fold higher than that in the rough micro- 
some fraction. The activity in the smooth 
microsome fraction was almost four times 
higher than that in the total microsomes. 
The smooth microsome fraction (Table V) 
shows a galactosyltransferase activity 
which indicates a 12% of contamination 
with Golgi complex, from the transferase 
activity found in Golgi complex-rich frac- 
tion 2 (Table V). Golgi complex-rich frac- 
tions 1 and 2 have the highest kallikrein 
activity. The combined Golgi complex 
fractions contain 71 and %% of the kalli- 
krein found in total microsomes and in 
smooth microsome fractions, respectively. 



TABLE IV. Distribution of Renal Esterase 

Activity in Different Components of the 

Microsomal Fraction of Salt-Loaded 

Colchicine-Treated Rats 



Fraction 



Esterase activity 



Homogenate 

Microsomes 

Rough microsomes 

Smooth microsomes 

Golgi complex-rich fraction I 

Golgi complex-rich fraction 2 

Supernatant 



0.145 ±0.016 
0.280 ± 0.019 
0.160 ±0.014 
0.970 ± 0.010 
1.800 ±0.150 
1.905 ±0.186 
0.025 ± 0.003 



Note. Values represent the mean ± SD of three prepara- 
tions obtained from 10 pooled, colchicine-pretreated, and 
sodium-loaded rat kidneys. Enzyme activity is expressed as 
ftmole BAEE hydrolyzed/min/mg of protein, at 3T*. 



In addition, the stimulating effect of the 
different fractions on uterine contractility 
were 100% sensitive to trasylol. 

A similar distribution of activities in all 
the mentioned fractions was obtained when 
the activity of esterase was determined, as 
is shown in Table IV. Again, in animals 
previously treated with colchicine, the two 
Golgi complex fractions show the highest 
level of kallikrein activity. 

Since the Golgi complex-rich fractions 
have low yields of proteins, both assays 
used to measure activity of kallikrein were 
not sufficiently sensitive to measure it in 
Golgi complex-rich fractions of nonloaded 
rats, in the absence of an enzymatic induc- 
tion. For this reason we could not measure 
it in these fractions. 

Since the amount of protein in the Golgi 
complex fractions 1 and 2 was rather low, 
both fractions were combined. The mor- 
phologies are shown in Fig. 4, and indicate 
that they derive predominantly from the 
Golgi complex. The Golgi complex fraction 
from colchicine-treated rats seems to con- 
tain more loaded Golgi complex elements. 

As shown in Table V, the Golgi com- 
plex-rich fractions 1 and 2, isolated from 
sodium-loaded rat kidneys, were enzy- 
matically unique compared to other pu- 
rified cell fractions (see Fig. 5). Thus, these 
fractions exhibit the highest level of activ- 
ity for galactosyltransferase, and appear 



22» 



GOLOf COMPLEX AND KALLIKREIN EXCRETIOK 



9^ 






"5 « 



4l! 



if 



*. A A -T ^ *■• — A 

— -r "T « « «^. ^» "^ 
"TLTL'" Ti :=: z 



;§; 



' 5 o 5 z « 

' '^. — s — ^1 



§; 



' *r. **-. 9 — r» 
. r*- *r. «r, ^ r» 
' r» r# O O O 



V 1 - »^ ^! «. 



« -» r<- o < 



yl 






^^ 



a. 



oooooooo 



••- S* * ^. O r^. 3C -» ri O 'n 
C g "^ — — — ri *^. r^, f*", — 



3C -» ri o 'n 






I] 



5C ac — ri ""T '^. "^ n 



X 






ti C C 



g 

3 IS I 



iiliriti 

2: -I -i y: oi O C cr. 



s 
jP 



3S 



^5 '-» 



jC 



? 






5f H. 






O C 

I E 

t c 



2^ E 






c: i« « o 



to be about 70 and 75^ pure^ respec- 
tively. 

The gliicose-6'pliospliaiase activity pres- 
ent in smooth microsomes shows that the 
Golgi complex firactions contain about 15 to 
2tfi endoplasmic reticuhun contamination. 
The acid phosphatase activity indicates that 
they are contaminated 3 to Kt with iyso- 
somes. Succinate cytochrome c mhictase 
activity suggests that they are also con- 
taminated n to 4%) with mitochondria. 
From the galactosyltransferase activity 
found in Golgi complex-rich fraction 2, it 
appears that smooth and the total micro- 
somal firactions are contaminated with the 
Golgi complex, about 9 and 12%, respec- 
tively. On the other hand, the level of acid 
phosphatase activity in the light mitochon- 
drial fraction appears to indicate a 28% 
contamination with lysosomes. 

The same marker enzymes, in purified 
subcellular fractions of sodium-loaded rats 
and sodium-loaded rats pretreated with low 
and high doses of colchicine were mea- 
sured, and the activities did not differ sig- 
nificantly from the values presented in 
Table V. In general, the values illustrated in 
this table are quite similar to the specific 
activity of the same marker enzymes of 
purified organelles isolated from untreated 
or from unloaded rat kidneys (29). 

Discussion. Due to previous evidence 
suggesting different sites of subcellular 
localization of kallikrein biosynthesis and 
secretion we focused our efforts on local- 
izing the organelles possibly involved in 
the biosynthesis and/or activation of renal 
kallikrein. In order to analyze the con- 
flicting evidence available on the subject, 
we have isolated different subcellular 
fractions from kidneys of rats submitted to 
two different treatments. Our purpose was 
to inhibit the secretion process and to ob- 
tain an increase in renal kallikrein content. 

As we reported previously (11) acute 
NaCl loading induces in the rat a rapid and 
considerable increase of renal kallikrein 
that could be due to the existence of a 
sodium receptor in the gastrointestinal tract 
or to a rapid change in the release mecha- 
nism of renal kallikrein following sodium 
absorption. 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



229 



. 7,i,'^'-:.^i 




• -y# ♦' ■*' m^ 



Fig. 5. Electron micrograph of a purified fraction obtained by fractionation of a kidney homogenate 
from a sodium-loaded rat. (a) Mitochondria-rich fraction, y 19,200; (b) smooth microsome-rich frac- 
tion, X 25,600: (c) rough microsome-rich fraction, x 29.600. 



Our present results demonstrate that have a high activity of kallikrein in the 

subcellular fractions, isolated by differen- crude light mitochondrial fraction. This 

tial centrifugation from sodium-loaded ani- fraction has a significant lysosomal con- 

mals not treated with colchicine (Table I), tamination as is indicated by its acid phos- 



230 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



phatase activity (Table IV). Based on the 
phosphatase activity, our results agree with 
those obtained by Carvalho and Diniz (6) 
and Baggio et al. (7). These authors postu- 
lated that lysosomes are the site where 
most of the kallikrein activity resides. 
However, results obtained by Chiang et al. 
(30) and Geipert and ErdOs (5) suggest that 
kallikrein is secreted into the extracellular 
space in the form of dense granules and 
probably through fusion with the plasma 
membrane. Our studies seem to indicate 
that this mechanism may also occur in kid- 
ney cells. When using colchicine, a secre- 
tion inhibitor that does not interfere with 
the protein transport from the rough endo- 
plasmic reticulum to the Golgi complex 
(13), the enzyme activity was recovered in 
the microsomal fraction, specifically in the 
Golgi complex-rich fraction. This indicates 
that the action of colchicine at the Golgi 
complex level may affect the secretion of 
granules containing kallikrein. Moreover, 
the acid phosphatase activity remains un- 
affected by colchicine treatment, indicating 
that lysosomal contamination in the light 
crude mitochondrial fraction of nontreated 
and colchicine-treated rats was similar. The 
fact that high kallikrein activity appears in 
the microsomal fraction after colchicine 
treatment is in agreement with results ob- 
tained by Nustad and Rubin (9). These au- 
thors have reported that kallikrein is mainly 
localized in this membranous fraction. Mi- 
crosomes, isolated according to our proce- 
dure (29), mostly include fragmented rough 
and smooth endoplasmic reticulum as well 
as Golgi complex membrane. For this rea- 
son, we associate the site of biosynthesis 
and/or activation of kallikrein with the 
Golgi complex. 

Studies by Redman (13) have shown that 
colchicine, administered to whole rat, 
blocks the secretion of serum albumin in rat 
liver cells blocking the release of this com- 
ponent from the Golgi complex. Our data 
suggest that colchicine administered to 
whole animal also blocks the release of kal- 
likrein from this organelle in kidney cells. 
Carvalho and Diniz (6) and Baggio et al. (7) 
reported a high level of kallikrein activity in 
the lysosomal fraction. We suggest that 



their results can be explained by the fact 
that kallikrein is secreted as secretory 
granules, which are physicochemically 
similar to the kidney lysosomes, and that 
after homogenation both sediment together. 

Renal kallikrein is a glycoprotein with a 
yet not determined carbohydrate sequence. 
Nevertheless, it is clear that glycoproteins 
are synthesized in the rough endoplasmic 
reticulum (peptidic backbone), and then 
transported to the Golgi complex (31). In 
this last organelle, sugars are added step- 
wise to the nonreducing end of the carbo- 
hydrate chains of the glycoprotein, by the 
specific action of galactosyl or syalyl 
transferase, using nucleotide sugar glyco- 
syl as donors (32). It is most likely that 
kallikrein is not only concentrated but also 
modified in the Golgi complex by the addi- 
tion of a terminal sugar. 

The evidence offered in this paper sup- 
ports the idea that kallikrein is transported 
within renal cells by a system of secretory 
vesicles derived from the Golgi complex. 
Our data also suggest that lysosomes may 
not play an important role in this process. 
On the other hand, since our data indicate 
that kallikrein activity is present in micro- 
somes, and particularly in the Golgi com- 
plex, the biosynthesis and/or activation of 
renal kallikrein seems to occur in these or- 
ganelles. 

Finally, our results provide a feasible ex- 
planation for the apparently conflicting re- 
sults on the sites of renal kallikrein biosyn- 
thesis and/or activation. 



The authors wish to acknowledge the competent 
technical assistance of Miss Maria I. Navarrete and 
Mr. Eddie Gonzalez. 

This work was supported by the Servicio de Desa- 
rrollo Cientffico, Artfstico y de Cooperation Interna- 
cional de ia Universidad de Chile (Grant B952-8233) 
and PNUD/UNESCO RLA Program 076/006 (Grant 12). 



1. Webster ME, Gilmore JP. Influence of kallikin-10 
on renal function. Amer J Physiol 206:714-718, 
1964. 

2. Carretero OA, Scicli AG. Renal killikrein its 
localization and possible role in renal function. 
Fed Proc 35:194-198. 1976. 

3. (Z)rstavik TB, Nustad K, Brandtzacg P, Pierce JV. 



GOLGI COMPLEX AND KALLIKREIN EXCRETION 



231 



Cellular origen of urinary kallikrein. H. His- 
tochemCytochcm 24:1037-1039, 1976. 

4. Nustad K, Va^ J. Pierce JV. Synthesis of kalli- 
krein by rats kidney slices. Brit J Pharmacol 
53:229-234. 1975. 

5. Geipert F, Erdos EG. Properties of granule that 
contain kallikrein and renin. Experientia 
27:912-913, 1971. 

6. Carvalho IF, Diniz GR. Kining forming enzyme 
(kininogenin) in homogenate of rat kidney. 
Biochim Biophys Acta 128:136-148, 1966. 

7. Baggio B, Favaro S, Antonello A, Zen F, Borsatti 
A. Subcellular localization of renin and kallikrein 
in rat kidney. Ital J Biochem 24:199-206, 1975. 

8. Nustad K. Localization of kininogenase in the rat 
kidney. Brit J Pharmacol 39:87-89, 1970. 

9. Nustad K, Rubin I. Subcellular localization of 
renin and kininogenase in the rat kidney. Brit J 
Pharmacol 40:326-333, 1970. 

10. Ward PE, Erdos EG, Gedney CD. Dowben RM, 
Reynolds RC. Isolation of renal membranes that 
contain kallikrein, angiotensin 1 -converting en- 
zyme (kininase (II) and angiotensinase in the rat. 
Clin Sci Mol Med 51:267s-270s. 1976. 

11. Rojas M, Brandan E. Zambrano F. Salt loading 
effect on renal kallikrein activity in rat. Proc Soc 
Exp Biol Med 163:151-154, 1980. 

12. Reaven E, Reaven G. Evidence that microtubules 
play a permissive role, in hepatocyte very low 
density lipoprotein secretion. J Cell Biol 
84:28-29. 1980. 

13. Redman CM, Bemeijee D, Howell D, Palade GE. 
Colchicine inhibition of plasma protein release 
from rat hepatocytes. J Cell Biol 66:42-59, 1975. 

14. Bamafi L, Rosas R, de la Lastra M, Croxatto HR. 
Influence of oxitocin and vasopresion on sodium- 
retaining activity of aldosterone in the rat. Amer J 
Physiol 198:255-259. 1960. 

15. Fleischer B. Zambrano F. Golgi apparatus of rat 
kidney. Preparation and role in sulfatide forma- 
tion. J Biol Chem 249:5995-6003, 1974. 

16. Stain Y, Widnell C, Stein O. Acylation of 
lysophosphatides by plasma membrane fractions 
of rat liver. J Cell Biol 39:185-192, 1968. 

17. Fleischer S, Kervina M. Subcellular fractionation 
of rat liver. In: Fleischer S, Packer L, eds. Meth- 
ods in Enzymology. New York, Academic Press, 
Vol. 31:pp6-41, 1974. 

18. Maunsbach AB. Isolation of kidney lisosomes. In: 
Fleischer S, Packer L, eds. Methods in Enzymol- 
ogy. New York. Academic Press, vol. 31:pp330- 
339, 1974. 

19. Dallner GP, Siekevitz P, Palade GE. Biogenesis of 



endoplasmic reticulum membranes in developing 
rat hepatocyte. J Cell Biol 30:97-115, 1966. 

20. Swanson MA. Glucose-6-phosphatase from liver. 
In: Colowick S, Kaplan NO. eds. Methods in En- 
zymology. New York. Academic Press, Vol. 
2:pp541-543, 195. 

21. Fleischer S, Fleischer B. Removal and binding of 
polar lipids in mitochondria and other membrane 
systems. In: Stabrook RW, Pullman ME, eds. 
Methods in Enzymology. New York, Academic 
Press, Vol. 10:pp406-433, 1%7. 

22. Besseys OA, Lowry OH, Brock MJ. A method for 
the rapid determination of alkaline phosphatase 
with five cubic millimeters of serum. J Biol Chem 
164:321-326. 1946. 

23. Chen PS, Toribara TY. Warner H. Microdetermi- 
nation of phosphorus. Anal Chem 28:1756-1758, 
1976. 

24. Fleischer B. Isolation and characterization of 
Golgi apparatus and membrane from rat liver. In: 
Fleischer S, Packer L. eds. Methods in Enzymol- 
ogy. New York, Academic Press. Vol. 31:ppl80- 
191, 1974. 

25. Croxatto HR, Noe E. Kallikrein-like enzyme in 
purified renal extracts containing renin. Comentat 
Pontif Acad Sci 40:1-2, 1971. 

26. Brown ME. The colorimetric determination of ar- 
ginine ester hydrolysis by human sera. J Lab Clin 
Med 55:616-624, 1960. 

27. Lowry OL, Rosebrough NJ, Farr AL, Randall RJ. 
Protein measurement with the Folin phenol re- 
agent. J Biol Chem 193:265-275, 1951. 

28. Fleischer B, Fleischer S, Osawa H. Isolation and 
characterization of Golgi membranes from bovine 
liver. J Cell Biol 43:59-79, 1%9. 

29. Fleischer B. Zambrano F. Fleischer S. Biochemi- 
cal characterization of the Golgi complex of 
mammalian cells. J Supramol Struct 2:737-750, 
1974. 

30. Chiang TS, Erdos EG, Miwa IM, Tague LL, Coal- 
son J. Isolation from salivary gland of granules 
containing renin and kallikrein. Cir Res 
23:507-517, 1968. 

31. Morre J. Outracht L. Dynamics of the Golgi ap- 
paratus. Membrane differentiation and membrane 
flow. Int Rev Cytol (SuppI 5):61-188, 1977. 

32. Fleischer B. The role of Golgi apparatus in the 
synthesis of glicoproteins and glycolipids. In: 
Harmon RE, ed. Cell Surface Carbohydrate 
Chemistry. New York, Academic Press, Vol. 
l:pp27-47, 1978. 

Received April 1. 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 232 (1982) 



Erratum 



Volume 171, No. 4 (1982) in the article, ** Annual Report of the Executive Sec- 
retary and Editor for the Year Ending December 31, 1981," pages 523-524, on 
page 524, the following lines were omitted and are to replace lines 10 and 11: 

The following editors will be retiring. They are: Drs. J. C. Beck, D. G. 
Gilmour, C. G. Harford, M. E. Lamm, C. C. Lushbaugh, R. J. Owellen, 
A. A. Spector, R. S. Speirs, and A. Stracher. 



Statement of ownership, management and circulation required by the Act of October 23, 1%2. Section 4369, Title 39, 
United States Code: of 

PROCEEDINGS OF THE SOCIETY 
FOR EXPERIMENTAL BIOLOGY AND MEDICINE 
Published monthly (except August) by Academic Press. Inc., Ill Fifth Avenue, New York. N.Y. 10003. Number of issues 
published annually: II. Editor: M. R. Nocenti. 630 West 168th St., New York. N.Y. 10032. Owned by Society for 
Experimental Biology and Medicine. 630 West 168th St.. New York, N.Y. 10032. Known bondholders, mortgagees, and 
other security holders owning or holding I percent or more of total amount of bonds, mortgages, and other securities: 
None. 

Paragraphs 2 and 3 include, in cases where the stockholder or security holder appears upon the books of the company as 
trustee or in any other fiduciary relation, the name of the person or corporation for whom such trustee is acting, also the 
statements in the two paragraphs show the affiant's full knowledge and belief as to the circumstances and conditions under 
which stockholders and security holders who do not appear upon the books of the company as trustees, hold stock and 
securities in a capacity other than that of a bona fide owner. Names and addresses of individuals who are stockholders of a 
corporation which itself is a stockholder of bonds, mortgages or other securities of the publishing corporation have been 
included in paragraphs 2 and 3 when the interests of such individuals are equivalent to I percent or more of the total amount 
of the stock or securities of the publishing corporation. 

Total no. copies printed: average no. copies each issue during preceding 12 months: 6460; single issue nearest to filing date: 
6235. Paid circulation (a) to term subscribers by mail carrier delivery or by other means: average no. copies each issue 
during preceding 12 months: 5711; single issue nearest to filing date: 5230. (b) Sales through agents, news dealers or 
otherwise: average no. copies each issue during preceding 12 months: 0; single issue nearest to filing date: 0. Free 
distribution by mail, carrier delivery, or by other means: average no. copies each issue during preceding 12 months: 154; 
single issue nearest to filing date: 154. Total no. copies distributed: average no. copies each issue during preceding 12 
months: 5865; single issue nearest to filing date: 5384. 

(Signed) M. R. Nocenti, Editor 

232 

Copyright (§) 1982 by the Society for Experimentsd Biology and Medicine. 
-lO rights reserved. 



^fUewU 0/ the Society 



Our members are requested to note the following list of Friends. We 
wish to express our thanks to each of them. Their contributions help 
materially in meeting the very sharp increase in costs of publication. 



FRIENDS 



Ciba Pharmaceutical Products, Inc. 

E. I. du Pont de Nenfiours & Co. 

Eli Lilly and Co. 

Hoffnfiann-LaRoche, Inc. 

Mead Johnson & Co. 

Merrell Dow Pharmaceuticals, Inc. 

Chas. Pfizer and Co., Inc. 

Ortho Pharmaceutical Corporation 

A. H. Robins Co. 

Rogoff Foundation 

Sandoz Pharmaceuticals 



Schering Corporation 
Sharp and Dohme, Division of Merck & Co., 
Smith Kline and French Laboratories 
Squibb Inst, for Medical Research 
Stuart Pharmaceuticals, Div. of ICI 

Americas, Inc. 
Upjohn Co. 

USV Pharmaceutical Corp. 
Wallace Laboratories, Division of Carter- 
Wallace, Inc. 
Wyeth Laboratories, Inc. 



Inc. 



NOTICE TO CONTRIBUTORS 

General Instructions 

Manuscripts should be written in clear, concise and grammatical English, and should conform to the general 
style of the Journal and the specific instructions listed below. Manuscripts which are not adequately prepared 
will be returned to the authors, since it is not feasible for the Editors to undertake extensive revision or rewriting 
of manuscripts submitted. Contributors, particularly those unfamiliar with English usage, are encouraged to seek 
the help of colleagues in the preparation and review of manuscripts prior to submission. This practice will usually 
reduce the time required for review and will avoid delays in the publications of the manuscript. 

SUBMIT MANUSCRIPTS IN TRIPLICATE (ONE ORIGINAL AND TWO COPIES). A $10.00 CHECK OR 
MONEY ORDER (NON-REFUNDABLE) MUST ACCOMPANY THE MS TO COVER HANDLING COSTS 
FOR ALL MSS RECEIVED. Submitted manuscripts and accompanying materials become the property of the 
Society, and the Society is not responsible for loss or damage to such materials. 

All manuscripts should be submitted to Dr. M. R. Nocenti, 630 W. 168th Street, New York, N.Y. 10032. 

EDITORIAL OFFICE IS CLOSED DURING AUGUST. 

Authors submitting manuscripts containing data from experiments involving recombinant DNA molecules 
must provide a statement for each of the two reviewers which certifies that their experiments complied with the 
NIH guidelines on physical and biological containment procedures. 

1. Only original papers will be considered. Manuscripts are accepted for review with the understanding that 
the same work has not been and will not be published nor is presently submitted elsewhere, and that all persons 
listed as authors have given their approval for the submission of the paper; further, that any person cited as a 
source of personal communications has approved such citation. Written authorization may be required at the 
Editor's discretion. Articles and any other material published in the Proceedings of the Society for Experimental 
Biology and Medicine represent the opinions of the author(s) and should not be construed to reflect the opinions 
of the Editor(s), the Society, or the Publisher. 

Authors submitting a manuscript do so on the understanding that if it is accepted for publication, copyright in 
the article, including the right to reproduce the article in all forms and media, shall be assigned exclusively to the 
Society. The Society will not refuse any reasonable request by the author for permission to reproduce any of his 
or her contributions to the journal. Send requests for permission to reproduce items published in Proceedings of 
the Society for Experimental Biology and Medicine to: Dr. Mero R. Nocenti, Editor, Society for Experimental 
Biology and Medicine, 630 W. 168th St., N.Y., N.Y. 10032. 

A manuscript rejected by the PSEBM should not be re-submitted. All manuscripts will be given a quality rating 
by the two reviewing editors; those manuscripts with low priority ratings will not be accepted even though they 
have been classed as generally acceptable. Split decisions will be decided on the basis of the two priority ratings 
and/or by a third review. 

2. a. Manuscripts should be as concise as possible, yet sufficiently detailed to permit critical appraisal. 

b. Manuscripts (including tables, legends, and footnotes) should be double or triple spaced. 

c. The first page of the manuscript should contain the complete title of the paper, category for the 'Table of 
Contents** (select from list in item 20), names of authors (without degrees), affiliations (including Zip 
Codes), and a running title consisting of no more than 40 characters (including spaces). The second page of 
the manuscript should give the name, complete address, and telephone number of the author to whom ALL 
correspondence should be sent. Please include Zip Code. 

d. Units of weights, measures, etc., when used in cor\junction with numerals, should be abbreviated and 
unpunctuated, e.g., 6 R, 3 g, 5 ml, 8% (see No. 21 below). 

3. Manuscripts of nonexperimental researches, or those with inadequate controls, are not acceptable. 

4. Unnecessary subdivision of a research into several manuscripts is not acceptable. 

5. a. Manuscripts devoted to improvement of procedure or of apparatus may be accepted when a new 
principle is involved or when decidedly superior biological results are obtained. Evidence of such superi- 
ority should be given. 

b. Confirmatory or negative results will not be accepted unless they are of obvious biological significance. 

6. Manuscripts should not exceed 40 typed pages (including tables, charts, and references). 

7. Title should be limited to 15 words. Manuscripts should contain an Abstract, Introduction, Materials and 
Methods, Results and Discussion. Abstracts should not exceed 300 words. 

8. Conclusions should be based upon experimental data submitted. 

9. Figures. All figures should be cited consecutively by Arabic numerals in the text with figure legends typed 
on a separate sheet. These should contain sufficient experimental detail to permit the flgure to be interpreted 
without reference to the text. Units should be clearly indicated in the figures themselves. Wherever possible, 
curves should be combined into a single figure in order to keep the number of illustrations to a minimum. 



PLEASE NOTE: All figures and illustrations are to be submitted in such form as to permit photographic 
reproduction without retouching or redrawing. This includes the lettering, which is reproduced as part of the 
photoengraving and is not set in type. Line drawings should be carefully drafted with black India ink on white 
drawing paper or blue drafting cloth, no larger than 8.5 x 11 .5 inches overall (21 x 27.5 cm). The lettering should 
be large enough to allow a reduction of two- thirds. High quality glossy prints are acceptable. 

10. Tables. These should be numbered with Roman numerals and cited consecutively in the text. Each table 
should be titled and typed double-spaced on a separate sheet. Refer to current issues of the Proceedings for the 
acceptable style of tables. The title of each table should cleariy indicate the nature of the contents, and sufficient 
experimental detail should be included in footnotes to the entries to permit the reader to interpret the results. 
Units must be clearly indicated for each of the entries in the table. To save space, repetition of similar experi- 
ments and columns which can be calculated from other entries in the table should be avoided wherever possible. 

11. Footnotes. Footnotes in the text should be identified by superscripts consisting of Arabic numerals and 
should be typed on separate sheet: footnotes in the tables should be identified with superscript lower-case letters 
a, b, c, etc., and placed at the bottom of the table. 

12. a. References. Only essential citations should be submitted, and they should be arranged numerically at 
the end of the manuscript. References to the literature should be cited in the text by Arabic numerals in 
parentheses, set on the text line. The references should be numbered consecutively in order of their 
citation. 

b. Abbreviations of journal titles should follow the style used in Chemical Abstracts (Vol. ACCESS. Key 
to the Source Literature of the Chemical Sciences, 1969 Edition). Please note the style of capitalization 
and punctuation for journal articles, books, and edited books in the following examples:. 

1. Ludens JH, Bach RR, Williamson HE. Characteristics of the antinatriuretic action of 
growth hormone. Proc Soc Exp Biol Med 130:1156-1158, 1969. 

2. Abramson DI. Circulation in the Extremities. New York, Academic Press, pOOO, 1%7. 

3. Newell A. Simon HA. Programs as theories of higher mental processes. In: Stacy RW, Waxman BD, 
eds. Computers in Biochemical Research. New York, Academic Press, Vol 2:p000, 1965. 

c. ''Personal communication," "unpublished,'* "submitted" and numerous abstracts should be excluded 
from the reference list. If the manuscript has been accepted for publication, include it in the reference list, 
giving journal, year, etc. If not accepted do not include it in the reference list. 

13. Trade or popular name or abbreviation of a chemical may be used only when preceded by the chemical or 
scientific name; thereafter, any of these names or abbreviations may be used. Trade names should begin with a 
capital letter. 

14. Structural formulas of chemicals should be used only when absolutely necessary. 

15. The Proceedings is copyrighted by the Society for Experimental Biology and Medicine. 

16. a. Authors are expected to discourage circulation of reprints for commercial purposes, 
b. Reprints are limited to 1300. 

17. Changes in galley proof should be absolutely minimal. Authors will be charged for excessive changes. 

18. If none of the authors is a member of the SEBM, the Ms must be sponsored by an SEBM member. The 
sponsor should write stating (a) how familiar he is with the author and his research, (b) whether the author is 
scientifically reliable, (c) that the manuscript conforms to Notice to Contributors, (d) that he has critically ex- 
amined the manuscript, (e) wherein the manuscript is a significant contribution to science, and (f) that he as- 
sumes financial responsibility involved. The sponsor should be in the same laboratory as the author. Where not 
of the same or recent past laboratory, he should specify why such letter comes finom a member at another laboratory. 

19. The authors are required to pay a part of the cost of publication in the form of a page charge of $20.00 per 
page. 

Notify the Executive Secretary at least one month before change of address. 

20. Biochemistry, Endocrinology, Growth/Development, Hematology, Microbiology/Immunology, Nutrition, 
Oncology, Pathology, Pharmacology, Physiology, Radiobiology, Tissue Culture. 

21. Abbreviations. Contributors are requested to use the following abbreviations: 



calorie 


cal 


centimeter 


cm 


counts per minute 


cpm 


cubic centimeter 


cm^ 


Curie 


Ci 


degree Celsius (Centigrade) 


-°C 


degree Fahrenheit 


op 



millimeter mm 

milliosmole mOsm 

minute min 

molal (concentration) m 

molar (concentration) M 

mole spell out 

molecular weight mol wt 



diameter 

gram 

hour 

inch 

inside diameter 

intramuscular 

intraperitoneal 

intravenous 

kilocalorie 

kUogram 

liter 

meter 

microliter 

micrometer 

milligram 

milliliter 



diam 

g 

hr 

in 

i.d. 

im 

ip 

iv 

kcal 

kg 

spell out 

m 

Ml 
fim 
mg 
ml 



nanogram 


ng 


nanometer 


nm 


normal (concentration) 


N 


osmole 


Osm 


ounce 


oz 


outside diameter 


o.d. 


parts per million 


ppm 


percent 


% 


picogram 


Pg 


revolutions per minute 


rpm 


second 


sec 


specific activity 


sp act 


square centimeter 


cm» 


square meter 


m» 


subcutaneous 


sc 


volt 


V 


volume 


vol 



Rapid Communications 

22. Rapid Communications must be typed on special forms to be compatible with the camera-ready method 
of reproduction. The forms are to be obtained from the Editorial Office. The manuscript must not exceed four 
special form pages in length, including text, illustrative material, tables, and references. The style of the man- 
uscript, including the references, must conform to the style specified in the ''Notice to Contributors** for reg- 
ular manuscripts (see above). The manuscript must contain an Abstract, Introduction, Materials and Methods, 
Results, and Discussion. Type single-spaced with 12 pitch letters. The original and two copies of the manu- 
script are to be mailed to the Editor. All improperly prepared manuscripts will be returned to the authors. 

23. All Tables, Figures, or glossy prints must be on separate sheets, and their legends must be typed on sep- 
arate sheets also. Do not insert these items into the text typed on the special forms. Follow instructions in item 9, 
above. 

24. Each copy of the manuscript must be accompanied by a letter in which the authors justify the request 
for rapid publication. The objective of Rapid Communications is to provide rapid dissemination of timely, 
significant, and novel data of special or general interest. The novel aspects of data must be described clearly 
in the manuscript. 

25. Manuscripts submitted for the Rapid Communications section will be reviewed expeditiously by a single 
reviewer. It is anticipated that such manuscripts will be in print two months after their receipt. 



Society for Experimental Biology and Medicine 

630 WEST 168 STREET. NEW YORK. N. Y. 10032 
SUBMIT IN DUPLICATE 

Name of nominee Date 

(Please print Usi name first) 

Mailing address Type of Nomination 

(Check one) 

Regular 

Associate 

Title now held Student 

Highest Degrees (give data and institution) 

Highest positions held in last 10 years in experimental biology and medicine, including post- 
doctoral fellowships (name of institution, title, dates) 



Honors, awards, etc. 



List membership in national scientific societies devoted to experimental biology and medicine 



The undersigned members of the Society agree that the above nominee is eligible for 

membership 

Signatures of 2 sponsoring members: 1 

2 



(over — for instructions) 



Eligibility 

Regular Membership: ''An applicant who has independently (not under direct super- 
vision of another scientist) published original meritorious investigations in experimental 
biology or experimental medicine . . . and who is actively engaged in experimental 
research may be elected to membership/' In general, applicants should be at least 2 years 
beyond their doctorate degree and beyond postdoctoral training. 

Associate Membership: ''Individuals who are engaged in research in experimental biol- 
ogy or medicine and/or teaching or administration in these areas. They must be sponsored 
by two regular members of the Society and must be residents of North America.*' 

Student Membership: "Any student who is a candidate for a doctorate degree and is 
actively engaged in research in experimental biology or medicine, as attested to by two 
regular members of the Society (only one of whom needs to write a letter, but both need to 
sign the application), and who is a resident of North America.'' 

Sponsorship, The application must be sponsored i.e., signed, by at least two members of 
the society. One detailed and factual letter from one of the member-sponsors describing 
the basis for eligibility must accompany this application. This letter should point out the 
contribution to science made by the applicant. Student applicants should have passed 
their qualifying examination and should be actively engaged in their thesis research proj- 
ect; date (month, year) on which doctorate degree is expected should be stated in the 
letter. 

Bibliography, Submit on separate sheets a complete list of publications, listing authors, 
title of publication in full, journal, year, volume and pages. Do not include articles not yet 
accepted for publication. Abstracts, reviews and books must be listed separately. 

Reprints, Submit one copy of each of three of your reprints. 

Address, Send this application in duplicate and supporting data to the Executive Secretary, 
Society for Experimental Biology and Medicine, 630 W. 168th Street, New York, N.Y. 
10032. Applicants will be notified of action taken as soon as the application has been pro- 
cessed (usually within a few months). 

Do not use this side of form. Use separate sheet if needed. 



Council 



President, Theodore Cooper '83 
The Upjohn Company 

President-Elect, Marjorie B. Zucker '83 
New York University 

Past-President, Robert W. Berliner '83 
Yale University 

Treasurer, Harriet P. Dustan '84 
University of Alabama 

Executive Secretary, Mero R. Nocenti {ex officio) 
Columbia University 



UMVERSmrOF 

mcHiOMi 

JAM19B83 

PHOEMl 
UBRARV 



Lloyd L. Anderson '83 
Iowa State Univ. 

R. AUERBACH '83 

Univ. of Wisconsin 

H. A. Bern '83 
Univ. of California 

Walter J. Bo '85 
Bowman-Gray Sch. of Med. 

M. G. Horning '83 
Baylor Coll. of Med. 



John H. Laragh '86 
Cornell Med. Ctr. 

Julian B. Marsh '85 
Med. Coll. Pennsylvania 

S. M. McCann '83 
Univ. of Texas 

John Resko '85 
Univ. of Oregon 

Wilbur H. Sawyer '86 
Columbia University 



Arthur A. Spector '86 
Univ. of Iowa 

J. Tepperman '83 

State Univ. of New York 

A. K. Weiss '83 
Univ. of Oklahoma 
Chairman, Membership Comm. 

M. B. Zucker '82 
New York Univ. 
Chairman, Publ. Comm. 



membership application 

Membership in the Society for Experimental Biology and Medicine is open to all individuals who 
have independently published original meritorious investigations in experimental biology or experi- 
mental medicine and who are actively engaged in experimental research. In general, applicants should 
be beyond a supervised postdoctoral experience in order to be able to demonstrate the ability to 
conduct independent investigations. 

Associate Membership is available for individuals who are engaged in research in experimental 
biology or medicine or in teaching or administration in these areas. Student Membership is available to 
any student who is a candidate for a doctorate degree and is actively engaged in research in experi- 
mental biology or medicine. Write to the Executive Secretary of the Society for details about eligibility 
requirements and application procedures for Associate and Student Memberships. 

Application forms may be obtained from the Office of the Executive Secretary, Society for Experi- 
mental Biology and Medicine, 630 W. 168th St., N.Y., N.Y. 10032. In addition, a tear-out applica- 
tion form is printed in each issue of the PSEBM. 



Proceedings of The Society for Experimental Biology and Medicine. Vol. 171, No. 3, December 1982. Published 
monthly except August by Academic Press, Inc.. Ill Fifth Avenue. New York, N.Y. 10003. Second class postage 
paid at New York. N.Y. and at additional mailing offices. 1982: Subscription per year $68.00 U.S.A.; $88.75 out- 
side U.S.A. 1983: Subscription per year $73.00 U.S.A. and Canada; $96.00 outside U.S.A. and Canada. All prices 
include postage and handling. Send notices of change of address to the Office of the Publisher at least 6-8 
weeks in advance. Please include both the old and new addresses. Postmaster: Send all changes of address to 
Proceedings of the Society for Experimental Biology and Medicine, 111 Fifth Avenue. New York, N.Y. 10003. 
Copyright © 1982 by the Society for Experimental Biology and Medicine. 



Proceedings 

of the 

Society 

for 

Experimental Biology and Medicine 

including the following sections 

District of Columbia Southeastern 
Southwestern 



October-December 



VOLUME 171 



New York 



CONTENTS 



SQENTinC PROCEEDINGS. VOLUME 171 

Seven hundred-thirtieth issue, October 1982 1 

Seven hundred-thirty-first issue, November 1982 127 

Seven hundred-thirty-second issue, December 1982 233 

Author Index 315 



Proceedings of The Society for Experimental Biology and Medicine, Vol. 171, No. 3, December 1982. Published 
monthly except August by Academic Press, Inc., Ill Fifth Avenue, New York, N.Y. 10003. Second class postage 
paid at New York, N.Y. and at additional mailing offices. 1982: Subscription per year $68.00 U.S.A.: $88.75 outside 
U.S.A. 1983: Subscription per year $73.00 U.S.A. and Canada; $96.00 outside U.S.A. and Canada. All prices 
include postage and handling. Send notices of change of address to the Office of the Publisher at least 6-8 weeks 
in advance. Please include both the old and new addresses. Copyright © 1982 by the Society for Experimental 
Biology and Medicine. 



TABLE OF CONTENTS 

BIOCHEMISTRY 

Characteristics of Rat Skull Benzylamine Oxidase T. H. An dree, D. E. Clarke 298 

ENDOCRINOLOGY 

17/3-Hydroxysteroid Dehydrogenase Activity in Tissues of J. A. Resko, H. L. Stadelman 233 

Fetal Rhesus Macaques 

Reserpine Inhibits Rat Anterior Pituitary Hormone Secretion I. S. Login, A. M. Judd, M. O. 

in Vitro: Effects on GH, TSH, and LH Thorner, R. M. MacLeod 247 

Effect of Prolactin on Galactose-Induced Cataractogenesis O. Gona, S.-C. J. Fu 285 

in the Rat 

MICROBIOLOGY/IMMUNOLOGY 

Classification of Hybridomas to Respiratory Syncytial Virus B. F. Fernie, P. J. Cote, Jr., J. L. 

Glycoproteins Gerin 266 

Androgen Suppression of Circulating Immune Complexes R. M. Coleman, N. J. Rencricca, 

and Enhanced Survival in Murine Malaria P. T. Fawcett, M. C. Veale, M. A. 

LoCoNTE 294 

Transplantation of Cultured Bursal Epithelium: A Histologic P. Thomas, M. Dustoor, G. Splitter, 

Study R. Hong 306 

NUTRITION 

The Effect of Dehydroepiandrosterone on Adipose Tissue M. P. Cleary, R. Seidenstat, R. H. 

Cellularity in Mice Tannen, A. G. Schwartz 276 

PATHOLOGY 

6-Keto-PGFto Synthesis in Diabetic Rat Aorta: Effect of H. Wey, M. T. R. Subbiah 251 

Substrate Concentration and Cholesterol Feeding 
Oligoclonal IgG in the Cerebrospinal Fluid of Guinea Pigs M. Iivanainen, B. Driscoll, J. Rich- 
with Experimental Allergic Encephalomyelitis ert, M. Leon, A. Chu, M. Kies, 

B. Brown, W. Wallen, D. Mad- 
den, J. Sever 272 

PHARMACOLOGY 

Electrophysiologic Effects of Verapamil on Cells Bordering S. S. Wong, R. J. Myerburg, J. S. 
Healed Myocardial Infarction in the Cat Cameron, K. Epstein, A. M. Ezrin, 

P. Kozlovskis, a. L. Bassett 289 

PHYSIOLOGY 

Responsiveness of Cerebral Osmoreceptors in the Anesthe- C. R. Wesley. L. J. Huffman, J. P. 

tized Dog Gii more 238 

Age- and Strain-Related Differences in Metabolic Response M. Kiang-Ulrich, S. M. Horvath . . . 242 

to Tyramine in Rats 

Effect of Hypertonic Sodium Chloride on Polyribosomes and C. N. Murty, B. Oliveros, H. Sid- 

Protein Synthesis of Kidneys of Rats ransky 258 

Author Index for Volume 171 315 

Cumulative Subject Index for Volumes 169- 171 317 

Const rruT ION and Bylaws D-5 

Membership Direc iory 

Directory of Active Members (alphabetical) D-I5 

Directory of Active Members (zip code) D-63 

Directory of Emeritus Members D-99 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 233-237 (1982) 



17i8-Hydroxysteroid Dehydrogenase Activity 
in Tissues of Fetal Rhesus Macaques (41503) 

JOHN A. RESKO> and HENRY L. STADELMAN 

Department of Physiology, Oregon Health Sciences University, Portland, Oregon 97201, and 
Oregon Regional Primate Research Center, Beaverton, Oregon 97006 



Abstract. 17i8-Hydroxysteroid dehydrogenase (17/3-HSDH) activity was measured in tis- 
sues (anterior pituitary, diencephalon, frontal cortex, liver, whole blood, and uterus) of fetal 
rhesus macaques on Days 80, 120, and 150 of gestation. The production of estrone (Ej) 
(quantified by radioimmunoassay) by an 900g supernatant incubated with excess substrate 
(estradiol- 17)8 [E2]) was used as an index of 17i8-HSDH activity. Over a 30-min incubation 
period E, was produced in a linear fashion by all the tissues that we studied. The pH 
optimum for this activity in the pituitary gland was —9. Boiling of the tissue for I min 
reduced its activity significantly. Significant changes in 17j8-HSDH activity during develop- 
ment were observed for two of the six tissues studied. Activity in fetal anterior pituitary 
glands was greatest on Day 80 of gestation but changed to significantly lower levels by Day 
120 {P < 0.01). This low level continued on Day 150 of gestation. The pattern of activity in 
the liver was different from that in the pituitary gland. High activity was found on Day 80, 
and this increased significantly by Day 120 {P < 0.05). The levels of activity observed on 
Day 150 were similar to those found on Day 120. Low levels of 17/3-HSDH activity were 
found in the central nervous system, whole blood, and uterus. In rhesus adults as in fetuses 
we have found relatively high levels of 17/3-HSDH activity in the anterior pituitary gland. 
The biological significance of changing levels of this activity during gestation is not known. 



Many tissues of the reproductive system 
are estrogen sensitive, but the roles of met- 
abolic enzymes in mediating estrogen ac- 
tion are not well understood. We do know, 
however, that estrogens such as estradiol- 
17)8 (E2) can be converted to weaker com- 
pounds such as estrone (Ei) by microsomal 
enzymes, the 17)8-hydroxy steroid dehydro- 
genases (17)8-HSDHs), in reproductive tis- 
sues (1). Estrone is less active than E2, ap- 
parently because it is less competitive for 
the estrogen receptor (2) and diffuses out of 
target cells more easily (3). Considering 
these two differences alone in the prop- 
erties of E, and E2, one can envision effec- 
tive modulation of the actions of E2 by its 
cellular conversion to Ej. In uterine tissue 
from women, the 17i8-HSDHs are regulated 
by progesterone and the activity of these 
enzymes varies with the stage of the repro- 
ductive cycle (3). Similar observations have 



' To whom all correspondence should be addressed 
at the Oregon Health Sciences University. 



been made in uteri (4) and pituitaries (5) 
from rhesus macaques, but the significance 
of these observations is not well under- 
stood. All of the above studies involved tis- 
sues from adult animals. Recently, we had 
the opportunity to obtain tissues from fetal 
monkeys at different times in gestation, and 
we determined the activity of the 17)8- 
HSDHs in these tissues. 

Materials and Methods. Animals and 
tissue preparations. Tissues were obtained 
from 12 rhesus macaque (Macaca mulatta) 
fetuses of known gestational ages (Days 80, 
120, and 150) that had been delivered by 
cesarean section. The three Day-80 and the 
three Day- 150 fetuses were intact females. 
The six Day- 120 fetuses comprised both 
males and females; some of these had been 
gonadectomized 3 weeks before and given 
testosterone or saline infusions for 6 hr. In 
this latter group we combined the data from 
the various treatment groups because no 
differences appeared evident from the fol- 
lowing comparisons. The 17)8-HSDH ac- 
tivity in anterior pituitary glands of two 



233 

0037-9721 I^UUQl'iV^V^^V.^^ 
Copyn^l © YSmb'j xYft VicvtVj tat %Kvarvm«fi\A'«>s*«« 



234 



FETAL DEHYDROGENASE 



female fetuses spayed in utero on Day 100 
of gestation and treated with testosterone 
for 21 days ranged from 220 to 238 ng Ej mg 
protein"' after a 30 min incubation. 
Pituitaries from four males similarly treated 
contained 207-363 (range) ng Ei mg pro- 
tein"*. In a 15-min incubation a pituitary 
gland of a female not treated with testoster- 
one contained 132 ng Ej mg protein"* 
whereas the activities of this enzyme in the 
pituitaries of two females treated with tes- 
tosterone were 130 and 154 ng Ej mg pro- 
tein"*, respectively. These preliminary ob- 
servations revealed no striking differences 
between treatments thereby providing the 
rationale for combining animals between 
treatments. At delivery the fetuses received 
cold saline infusions to remove blood from 
tissues to be analyzed. The brains and pi- 
tuitaries were quickly removed and chilled 
with ice. Under these conditions the vari- 
ous tissues were dissected, weighed, placed 
in ice-cold buffer, and homogenized. The 
following tissues were removed and ana- 
lyzed for 17i8-HSDH activity: liver, uterus, 
anterior pituitary, frontal cortex, and di- 
encephalon (a small block of tissue between 
the optic chiasm and the mammillary bod- 
ies about 2 mm thick). Each tissue was 
homogenized in 1 ml of ice-cold tris(hy- 
droxymethyDaminomethane (Tris) buffer, 
pH 8.0, at 37°C in a 5-ml glass homogenizer. 
The homogenate was poured into a 13 x 
100-mm glass tube and centrifuged at 800^ 
in a Sorvall refrigerated centrifuge. An ali- 
quot of the supernatant was removed for 
Lowry protein determinations. An aliquot 
of 1.5 mg of protein from the 800^ super- 
natant was incubated in 0.75 ml Tris buffer 
with different quantities of Eg and with a 10- 
fold molar excess of NAD for different peri- 
ods of time. Each incubation flask contained 
1 .5 mg of protein, an amount we had already 
demonstrated to be on the linear portion of 
the activity versus protein concentration 
curve. 

Chrofyuitography. After incubation the 
steroid was extracted from the incubation 
medium in 6 ml of ether. The ether extract 
was taken to dryness under a stream of air 
in a water bath at 37''C. The samples, dis- 



solved in a small volume of chloroform: 
methanol (1:1, v/v), were applied to the 
origins of paper strips with capillary tubes. 
We used Whatman No. 1 paper strips (55 
cm long and 2.5 cm wide) that had been 
cleaned by being boiled in methanol. The 
chromatograms were developed in benzene: 
formamide. Benzene was the mobile phase, 
and the paper was soaked in formamide: 
acetone (1:1, v/v). In this system it takes 
approximately 3 h for the solvent front to 
reach the end of the paper. Ej and Ej (/?/S of 
0.33 and 0.71, respectively) clearly separate 
in this chromatography system). After chro- 
matography, the paper strips were dried at 
room temperature for 16 hr. Standard solu- 
tions of El and Ej fractionated with the 
samples on individual paper strips were used 
as references for elution. The standards 
were visualized on paper by immersion of 
the paper strips in Barton's reagent, i.e., 
Wc K3Fe(Cu)6and 1% FeCl3-6HjO(l:l, v/v) 
(6). An area with the mobility of Ei was 
eluted with 10 ml of methanol from each 
paper chromatogram. The methanol extract 
was taken to dryness and dissolved in 1 ml 
of ethanol. One-tenth of the ethanol extract 
was dried, dissolved in 100 yA of hexane: 
benzene: methanol (62:20:13, v/v), applied to 
a Sephadex LH-20 column, and fractionated 
in the same solvent system mentioned above. 
The E, area was collected from the Sepha- 
dex columns and quantified by radioimmu- 
noassay (RIA). We estimated procedural 
losses by adding 10,000 cpm of pH]estrone 
(E,) to duplicate samples which we carried 
through the entire procedure. Using these 
independent estimates of recovery, we ad- 
justed the fmal quantity of Ej in each sam- 
ple for procedural losses. Although E, was 
quantified by an antiserum that cross-reacts 
with Eg, we achieved specificity in our Ej 
assay by chromatography on paper and 
Sephadex LH-20 columns. This was demon- 
strated by the low blanks for E, (<10 pg) 
when no tissue, only substrate, was added. 
These blanks were subtracted in the process 
of computing the E, values. The measure- 
ment of E, was used as an index of 17)8- 
HSDH activity. 
Statistics. Differences between the mean 



FETAL DEHYDROGENASE 



235 



production of Ei among gestational ages 
were analyzed by / tests after the homoge- 
neity of variance had been determined. 

Results. In a previous publication we 
validated our use of RIA measurements of 
E, as an assay for 17)8-HSDH activity in 
tissues from adult rhesus macaques (S). 
Similar information on fetal tissues is pre- 
sented in Figures 1 through 3. Assay of the 
El produced from Eg by the 8(% superna- 
tant of fetal pituitaries and liver revealed a 
linear response up to minute 30 of incuba- 
tion, the last time period mentioned in Fig. 
1. Measurements of E^ in the 80Q^ super- 
natant at time revealed small amounts of 
E, in the tissue before the incubation had 
begun. These amounts were subtracted 
from the IS- and 30-min values; and from all 
other data presented in this manuscript. 
Similar responses for other tissues such as 
uterus, cortex, and whole blood were ob- 
tained, but less El was formed by these tis- 
sues in vitro. Figure 2 demonstrates the ef- 
fect of increasing substrate concentrations 
on 17)8-HSDH activity after a 15-min incu- 
bation. It appears that 80 fjM E2 the limit of 
solubility in our buffer is a near saturating 
quantity of substrate. The effects of pH and 
heat treatment (boiling for 1 min) are shown 
in Fig. 3. The pH maximum was about pH 
9.0 since the activity declined rapidly at pH 
10.0. These data generated for liver may or 
may not be applicable to other tissues. 
Boiling the 800^ supernatant for 1 min 




15 30 

IncubotKsn Time (mm) 

Fig. 1. Effects of incubation time on 17/3- 
hydroxysteroid dehydrogenase (17/3-HSDH) activity 
of liver and anterior pituitary glands from Day- 120 
fetuses in =6). 




so no 

pM (EJ 

Fig. 2. Effects of substrate concentration (estradi- 
ol- 17/3 [E2I) on 17/3-hydroxysteroid dehydrogenase 
(17i8-HSDH) activity in Day-120 fetal pituitaries. 



drastically reduced the capacity of the tis- 
sue to convert Eg ^ E, but some activity 
remained. This activity was not subtracted 
from any of the values reported in this 
manuscript. 

We used this technique to analyze the 
17i8-HSDH activity in various tissues from 
fetuses as a function of stage of gestation 
(Fig. 4). The greatest activities occurred in 
fetal livers and anterior pituitary glands. 
Activity levels in other tissues were much 
lower. Pituitary glands from Day-80 fetuses 
contained significantly more enzyme activ- 
ity than pituitary tissues taken at later times 
in gestation (P < 0.01). Activity in liver, 
however, increased significantly from Day 
80 to Day 120 of gestation (P < 0.05). Ac- 
tivities of all other tissues were low and did 
not seem to change as gestation progressed. 

Discussion. An understanding of the 
function of metabolic enzymes for promot- 
ing hormone action in target tissues is be- 
ginning to emerge. For target tissues such 






?2. 





uT 



ao M 

pH 

Fig. 3. Effects of pH and heat treatment on 17/3- 
hydroxysteroid dehydrogenase (17i8-HSDH) activity 
in liver from Day-120 fetuses. 



236 



FETAL DEHYDROGENASE 




(••3) 



(••5 



(it-e) 

Day of GestaHon 
Fig. 4. Effects of age on 17/3-hydroxy steroid dehy- 
drogenase (17/3-HSDH) activity in fetal tissues from 
rhesus macaques. Tissues were incubated with 80 /LtA# 
E, and an excess of NAD for 30 min. We used the 
amount of Ei produced by this extract as an index of 
17i8-HSDH activity. E, was quantified by a specific 
RIA after chromatography on paper and LH-20 col- 
umns. See Methods section for more details about 
procedures. Bars with the same letters differed signifi- 
cantly iP < 0.01 for a, b and P < 0.05 for c, d). 

as prostate (7) and perineal skin (8), pro- 
duction of dihydrotestosterone (DHT) from 
testosterone appears to be an essential con- 
version for androgen action. Likewise, the 
aromatization of testosterone to E2 in 
neural tissue appears to be important in 
many species for sexual differentiation of 
the brain and negative feedback control of 
gonadotropin secretion by steroids (9, 10). 
The 17i8-HSDHs are found in a number of 
mammalian tissues such as erythrocytes 
(11), uterine endometrium (1), liver (12), 
and placenta (13). The amounts of 17)8- 
HSDH activity in human endometrium in- 
crease during the secretory phase of the 
cycle and in response to exogenous pro- 
gesterone (14). Similar results have been 
obtained in uterine tissue from rhesus 
macaques (4). In that study both Eg and 
progesterone appeared to increase the ac- 
tivity of these enzymes. In addition to those 
in the uterus, we found high levels of 17/3- 
HSDH activity in adult anterior pituitary 
glands (5). Data on fetal pituitaries reported 
in this paper confirm this fact, but only with 



respect to fetuses in early gestation. Some 
reports link the production of DHT and E2 
from testosterone to the cellular action of 
this hormone in certain tissues, but the 
significance of the biological activity of 
the 17i8-HSDHs is more difficult to under- 
stand. In uterine endometrium from rhesus 
macaques the favored conversion is Eg —> 
El rather than vice versa (4). The produc- 
tion of E, from Ej can be viewed as simply a 
breakdown process or it can be construed 
as a more significant and dynamic way of 
controlling E2 levels within the target cell. 
If one takes the latter point of view, the 
effectiveness of E2 in initiating its biological 
action is proportional to the amount of ac- 
tive hormone that binds to the estrogen re- 
ceptor. Therefore, study of the production 
and control of 17)8-HSDH activity seems 
important to an understanding of the way in 
which E2 acts on target cells. 

The connection between the presence of 
significant amounts of 17)8-HSDH activity 
and a significant biological function is the 
unknown factor at this time. There is good 
evidence in some rodent species that 
aromatization of androgen to estrogen is 
important for sexual differentiation of the 
fetus. Our measurements of E2 in the serum 
of the fetus have shown very low or unde- 
tectable quantities of this hormone early in 
gestation, when differentiation of the ner- 
vous system and anlagen of the reproduc- 
tive tract occurs (15). Concentrations of E2 
in the maternal serum, however, were rela- 
tively high. Apparently, the primate fetus 
develops a mechanism to maintain low 
levels of E2 within itself and thus prevent 
abnormalities in sexual differentiation. The 
question is: Do 17)8-HSDH activities in the 
liver and pituitary have a significant role in 
this process? There are analogous situa- 
tions in which the metabolic conversion of 
steroid hormones correlates with physio- 
logical effects previously unexplained. 
Doves castrated several weeks earlier are 
insensitive to androgens that mediate perch 
calling (16). This insensitivity correlates 
with an increase in 5/3-reductase activity in 
the preoptic region of the brain (16). There 
are precedents in the literature for the idea 
that enzymes involved in the metabolism of 



FETAL DEHYDROGENASE 



237 



steroids to inactive products serve impor- 
tant biological functions. The fact that their 
activities can be regulated by compounds 
such as progesterone or E2 and the fact that 
they can be affected by gonadectomy indi- 
cate their importance in the physiological 
regulation of cells. 

This work was supported by Grants RR-00163 and 
HD-16022 from the National Institutes of Health. Pub- 
lication No. 1 174 of the Oregon Regional Primate Re- 
search Center. 

1. Tseng L, Gurpide E. Estradiol and 20a- 
dihydroprogesterone dehydrogenase activities in 
human endometrium during the menstrual cycle. 
Endocrinology 94:419-423, 1974. 

2. Tseng L, Gurpide E. Effect of estrone and pro- 
gesterone on the nuclear uptake of estradiol by 
slices of human endometrium. Endocrinology 
93:245-248. 1973. 

3. Tseng L, Gurpide E. Changes in the in vitro me- 
tabolism of estradiol by human endometrium 
during the menstrual cycle. Amer J Obstet 
Gynecol 114:1002-1108, 1972. 

4. Kreitman O, Kreitman-Gimbal B, Hodgen GD. 
17-/3- Hydroxy steroid dehydrogenase in monkey 
endometrium: Characterization of enzyme activ- 
ity and effects of estradiol alone or in combination 
with progesterone. Steroids 34(6):693-703, 1979. 

5. Resko JA, Stadelman HL, Norman RL. 17)8- 
Hydroxysteroid dehydrogenase activity in the 
pituitary gland and neural tissue of rhesus mon- 
keys. J Steroid Biochem 11:1429-1434. 1979. 

6. Barton GM, Evans RS, Gardner JAF. Paper 
chromatography of phenolic substances. Nature 
(London) 170:249-250, 1952. 

7. Wilson JD, Gloyna RE. The intranuclear metabo- 
lism of testosterone in the accessory organs of 



reproduction. Recent Prog Horm Res 26:309- 
336, 1970. 

8. Northcutt RE, Island DP, Little GW. An explana- 
tion for the target organ unresponsiveness to tes- 
tosterone in the testicular feminization syndrome. 
J Clin Endocrinol Metab 29:422-425, 1%9. 

9. Mainwaring WIP. The mechanism of action of an- 
drogen. Monogr Endocrinol, No. 10, New York, 
Springer Veriag. p35, 1977. 

10. Resko JA, Quadri SK, Spies HG. Negative feed- 
back control of gonadotropins in male rhesus 
monkeys: Effects of time after castration and in- 
teractions of testosterone and estradiol- 17)8. En- 
docrinology 101:215-224, 1977. 

11. Migeon CJ, Lescure WH, Zinkham WH, Sidbury 
JB Jr. In vitro interconversion of 16-C"-estrone 
and 16-C'*-estradiol-17/3 by erythrocytes from 
normal subjects and from subjects with a defi- 
ciency of red-cell glucose-6-phosphate dehydro- 
genase activity. J Clin Invest 41:2025-2053, 1%2. 

12. Baird DT. Synthesis and secretion of steroid hor- 
mones by the ovary in vivo. In: Zuckerman S, 
Weir BJ, eds. The Ovary. New York, Academic 
Press, Vol III:p305, 1977. 

13. Engel LL, Groman EV. Human placental 17/3- 
estradiol dehydrogenase: Characterization and 
structural studies. Recent Prog Horm Res 
30:139-169, 1974. 

14. Tseng L, Gurpide E. Induction of human endo- 
metrial estradiol dehydrogenation by progestins. 
Endocrinology 97:825-833, 1975. 

15. Resko J A, Ploem JG, Stadelman HL. Estrogen in 
fetal and maternal plasma of the rhesus monkey. 
Endocrinology 97:425-430, 1975. 

16. Hutchison JB, Steiner T. Brain 5/3-reductase: A 
correlate of behavioral sensitivity to androgen. 
Science 213:244-246, I98I. 



Received February 12, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 238-241 (1962) 

Responsiveness of Cerebral Osmoreceptors in the Anesthetized Dog (41504) 
C. R. WESLEY, L. J. HUFFMAN,* and J. P. GILMORE* 

Department of Physiology and Biophysics, University of Nebraska College of Medicine, 

Omaha, Nebraska 68105 



Abstract. Experiments were carried out to determine if, in the anesthetized dog under- 
going a water diuresis, selective elevation of cerebral osmolality induced an antidiuresis. 
Bilateral intracarotid infusion of hypertonic sodium chloride in an amount previously shown 
to increase jugular plasma osmolality by 3% was associated with a significant decline in 
^H,o* The same volume of hypertonic sodium chloride infused intravenously did not alter 
Ch o significantly. These experiments do not support the view that anesthesia blocks the 
response of cerebral osmoreceptors. 



The concept that cerebral osmoreceptors 
modulated the secretion of antidiuretic 
hormone (ADH) was suggested by Veraey 
as the result of studies done in the 1940s (1). 
He observed that intracarotid bolus injec- 
tions of hypertonic NaCl inhibited a water 
diuresis in the conscious dog. Since the 
antidiuretic response could be mimicked by 
the iiyection of posterior pituitary extract, 
Verney suggested that elevation of cerebral 
osmolality caused shrinkage of osmore- 
ceptor cells which, in turn, led to the re- 
lease of ADH. 

These studies were not seriously chal- 
lenged until the work of Bie. He found that 
in anesthetized dogs, elevation of cerebral 
osmolality by an estimated 5% had no effect 
on a water diuresis (2). In addition, the time 
course and extent of the antidiuresis pro- 
duced by an ic or iv 60-min infusion of a 
hypertonic NaCl solution was the same. 
Although at this time Bie did not deny the 
possibility that Verney was correct, he 
questioned if the response observed by 
Verney was specific. 

Bie has provided an in-depth exhaustive 
review of the concept of osmoreceptors, in 
which he reaffirms his position that the 
stimuli used in osmoreception studies have 



' lYcscnt address: IVparimont of Physiology. West 
Virginia rni\crsity School of Medicine. Morgantown. 
W Va :^5(>^ 

■ To N\ horn all correspi>ndence should be addressed. 



not been shown to be physiological osmo- 
stimuli (3). However, a very important part 
of the review is provided as an addendum in 
which a study by Dietz et al. (4) is dis- 
cussed. This group observed that in the 
conscious dog, a bilateral ic, 10-min infu- 
sion of hypertonic NaCl that was shown to 
increase jugular osmolality by only 3%, in- 
hibited a water diuresis. Since Dietz et aL 
used conscious dogs and Bie anesthetized 
dogs, Bie suggested that his failure to ob- 
tain results consistent with a central os- 
moreceptor hypothesis was due to the fact 
that he used anesthetized dogs while Ver- 
ney and Dietz et al, used conscious dogs, 
i.e., anesthesia blocks the cerebral os- 
moreceptor mechanism. To test this hy- 
pothesis, we carried out experiments in a 
group of anesthetized dogs, including those 
used by Dietz et aL, to determine if, when 
anesthetized with the same anesthesia used 
by Bie, they would show an antidiuresis 
when cerebral osmolality is elevated within 
a physiologic range. Contrary to Bie's hy- 
pothesis, the results show that anesthesia 
does not block the cerebral osmoreceptor 
mechanism. 

Methods. The study was conducted 
using a group of female American Fox- 
hounds weighing between 20 and 27 kg 
which had been prepared with bilateral ca- 
rotid loops. The dogs were fasted for 18 hr 
before an experiment. 

They were anesthetized with 60 mg/kg 
c¥-chloralose (Sigma Chemical) and 10 



238 



'Vi/Tyv'AL' //(LVS(J4S0I.(H)0 



nfitt r- /M(r.'A» th<^ .v«v«-f» av txrcnmtnisl BK4i^\ And Mc^Kine 



CNS OSMORECEPTORS IN ANESTHETIZED DOGS 



239 



mg/kg sodium pentobarbital (Nembutal, 
Abbott) administered via an intracath 
placed in a saphenous vein and advanced to 
the level of the right heart. A Foley catheter 
was placed in the bladder via the urethra 
and the carotids were catheterized with 
2-in., 18-gauge angiocaths. Blood pressure 
and heart rate were monitored via one ca- 
rotid line except during the bilateral IC in- 
fusion. 

A steady state water diuresis was estab- 
lished by infusing a volume of hypotonic 
solution of 25 mAf urea and 40 mAf glucose 
equivalent to 2% of body weight over a pe- 
riod of one hour. To maintain the diuresis, 
the urine output in any given period was 
matched by infusing an equal amount of 
glucose -urea solution. 

When a steady state urine flow rate was 
achieved, a 10-min, bilateral ic or a 10-min 
iv infusion of hypertonic NaCl was per- 
formed. The NaCl solution was infused at a 
rate of 45 /xA//kg body weight/min/artery or 
90 fiM /kg/min intravenously. Blood sam- 
ples for electrolytes, osmolality, creatinine 
concentration, and, in some experiments, 
arginine vasopressin (ADH) radioimmuno- 



assay (6) were taken immediately before 
and 9 min into the infusion period, and again 
at 60 min after the hypertonic infusion. At 
the end of the experiments, the animals 
were returned to their kennel to recover 
from the anesthesia. 

Data were analyzed using one-way 
analysis of variance for repeated measures. 
Significance reported is P < 0.05. 

Results. In Ave dogs given 90 /xA//min/kg 
NaCl intravenously (Fig. IB), no significant 
change occurred from the mean control 
creatinine clearance (GFR) of 52 ml/min. 
Free water clearance (Chjo) (control mean 
= 2.2 ml/min), urinary sodium excretion 
(UNaV) (control = 13.7 /leq/niin), and uri- 
nary potassium excretion (UKV) (control = 
8.1 fjLtq/min) did not change significantly 
from control values. Heart rate (HR) and 
blood pressure (BP) did not change. Plasma 
sodium concentration increased from a 
control of 141 meq/liter to 143 meq/liter 
after the hypertonic NaCl infusion and re- 
mained elevated for the duration of the ex- 
periment. Plasma osmolality increased 
from 290 to 294 mOsm/liter. In the three 
experiments in which it was measured, 



60 
CCR 

ni/win 40 
20 



UNoV 
j£q/min 



45 

30 

15' 



2-0 
CN(0 

mi/min '*' 

1.0- 

0.5 

O 

3.0 
Cotm 
tH/nm 2.0- 

1.0- 



-30 -S>0 -lOr^ 20 30 40 50 60-30 -2.0 -tOr^ 20 30 4Q SO 60 



■I I * I I 



l-H— f 



-+-.+-^^ 




■ — ' — I— • — I » — • — >- 



I— ^— »- 



-^^ 




Fig. 1. Results of ic (Panel A) and iv (Panel B) infusions of hypertonic NaCl solution. €„* creatinine 
clearance: UNaV, urinary sodium excretion: Ch,o. free water clearance: C^,,,. urinan osmolar clear- 
ance. Asterisks indicate statistical significance. P ^ 0.05. 



240 



CNS OSMORECEPTORS IN ANESTHETIZED DOGS 



ADH concentration remained less than 1.2S 
pg/ml. 

The response to bilateral ic infusion of 
hypertonic NaCl in eight dogs is shown in 
Fig. lA. Creatinine clearance remained 
near the control value of 42 ml/min 
throughout the experiments. UNaV in- 
creased significantly by 30 min after the be- 
ginning of the ic infusion, to a peak value of 
28.8 /Lieq/min from a control level of 12.5 
/Lieq/min. Chjo decreased significantly by 30 
min after the beginning of the ic infusion 
from a control level of 1.7 ml/min to a 
minimum of 0.4 ml/min indicating a pro- 
nounced antidiuretic effect of the ic infu- 
sion. No significant change occurred from a 
control UKV of 11.3 /xeq/min, or in HR or 
BP. Plasma sodium concentration showed 
precisely the same increase as with the iv 
infusion, that is, from 141 meq/liter to 143 
meq/liter. Plasma osmolality increased 
from 288 to 292 mOsm/liter. In the six ex- 
periments in which it was measured, ADH 
concentration remained less than 1.2S 
pg/ml. 

Discussion. The results clearly show that 
selective increases in cerebral osmolality in 
the chloralose anesthetized dog is as- 
sociated with a significant antidiuresis. 
These results are consistent with the results 
of Verney (1) and Dietz et aL (4) in the con- 
scious dog but inconsistent with the results 
of Bie (2, 3) in the anesthetized dog. Bie 
infused 83 /xA//kg/min ic, bilaterally for 8 
min and observed an increase in free water 
despite the fact that by the end of the infu- 
sion, peripheral venous osmolality had in- 
creased by approximately 7 mOsm/kg. The 
increase (4 mOsm/liter) in peripheral os- 
molality in the present experiments caused 
by an ic infusion was associated with an 
antidiuresis. However, the peripheral 
plasma osmolality in Bie's experiments was 
283.6 mOsm/liter while in the present ex- 
periments, it was 290 (iv) and 288 (ic) prior 
to the hypertonic sodium chloride infusion. 

Using dehydration as a stimulus in con- 
scious dogs, Robertson (5) found that the 
threshold A^„sm for ADH release was 286 
mOsm/liter while we found under these 
conditions a value of 297 mOsm/liter 
(Huffman and Gilmore, unpublished obser- 



vations). Studies in man and rats indicate 
that under conditions of hypervolemia (as 
in the present experiments and those of Bie 
due to the induction of a H2O diuresis), the 
osmotic threshold for ADH release would 
be increased. It is, therefore, quite possible 
that in the Bie experiments, because of the 
low resting Posm (283.6), the hypertonic 
NaCl infusions did not cause P^^^ to reach 
the osmotic threshold for ADH release 
while in the present experiments, because 
of the higher P^^^ threshold was reached 
and thus an antidiuresis ensued. The reason 
why the resting Pogm in Bie's experiments in 
which he infused hypertonic NaCl for 8 min 
was lower than in the present experiments 
is probably because he produced a 4% body 
weight hydration while we produced a 2% 
body weight hydration. Again, this greater 
state of hydration (hypervolemia) would be 
expected to produce a greater increase in 
the osmotic threshold for ADH release. The 
diuresis observed by Bie might be ex- 
plained as the result of increased fluid de- 
livery to distal nephron sites resulting in an 
increased generation of Ch„o- 

A major obstacle to conclusive results in 
studies such as these is the difficulty in 
measuring changes in ADH. Because of the 
hydration, even the relatively moderate 2% 
of body weight used in the present study, 
ADH is suppressed below concentrations at 
which changes from steady state, control 
concentrations can be measured accu- 
rately. From the ADH data reported under 
Results, it can be concluded that the 2% 
hydration was sufficient to suppress ADH 
release in response to the hydration. 
Whether any increase in ADH occurred due 
to ic or iv hypertonic NaCl infusion cannot 
be determined; it can only be stated that 
any increase was not sufficient to raise 
plasma ADH above 1.25 pg/ml, the lower 
limit of the assay used (6). 

In summary, the present study demon- 
strates the responsiveness of cerebral os- 
moreceptors during chloralose/pentobar- 
bital anesthesia, supporting the cerebral 
osmoreceptor concept of Verney (1). 

The authors wish to acknowledge the technical as- 
sistance of P. Anding, J. Durso. and W. Roccaforte. 
The manuscript was typed by R. Cozette. 



CNS OSMORECEPTORS IN ANESTHETIZED DOGS 



241 



1. Vemey EB. The antidiuretic hormone and the 
factors which determine its release. Proc R Soc 
London Ser B 135:25, 1947- 1948. 

2. Bie P. Studies of cerebral osmoreceptors in anes- 
thetized dogs: The effect of intravenous and intra- 
carotid infusion of hyper-osmolar sodium chloride 
solutions during sustained water diuresis. Acta 
Physiol Scand 96:306, 1976. 

3. Bie P. Osmoreceptors, vasopressin, and control of 
renal water excretion. Physiol Rev 60:%1, 1980. 

4. Dietz HR, Bie P, Gilmore JP, Share L, Zucker IH. 
The relation between carotid solute concentration 
and renal water excretion in conscious dogs. Acta 
Physiol Scand 114:45. 1982. 



Robertson GL. Athar S. Shelton RL. Osmotic 
control of vasopressin function. In: Andreoli TE, 
Grantham JJ, Recto FC Jr. eds. Disturbances in 
Body Fluid Osmolality. Bethesda, Md, Amer 
Physiol Soc. pI30, 1977. 

Cornish KG. Gilmore J P. Increased left atrial 
pressure does not alter renal function in the con- 
scious primate. Amer J Physiol (Reg Integrative 
Comp Physiol 12). RI19-R124. 1982. 



Received June 29. 1982. P.S.E.B.M. 1982. Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 242-246 (1982) 

Age- and Strain-Related Differences in Metabolic Response to Tyramine 

in Rats (41505) 

M. KIANG-ULRICH and S. M. HORVATH* 

Institute of Environmental Stress, University of California, Santa Barbara, Santa Barbara, California 93 106 



Abstract. Differences in metabolic response to intraperitoneally ii\jected tyramine were 
examined in young (3-4 month) and old (20-24 month) Sprague-Dawley and Fischer 344 
rats. Fischer 344 rats were found to be significantly more sensitive to tyramine than 
Sprague-Dawley. Optimal doses were 2 mg/kg body weight for young F344, 20 mg/kg for 
young Sprague-Dawley, 5 mg/kg for old F344, and >40 mg/kg for old Sprague-Dawley. 
Significantly higher doses were required by old rats than young. Mechanisms of these age- 
and strain-related differences in response are discussed. This information could well be 
important in selecting appropriate ages and strains of animals for experimental use, espe- 
cially in aging research. 



Tyramine, a naturally occurring sym- 
pathomimetic amine, has been identified as 
a normal constituent of some mammaUan 
tissues (1). Exogenously administered 
tyramine is actively taken up by the sym- 
pathetic nerve endings into the catechol- 
amine storage vesicles displacing catechol- 
amines, primarily norepinephrine (NE), in 
the process (2). There are no apparent signs 
of toxicity when tyramine is employed, and 
the elevated metabolism following injec- 
tions of either tyramine or NE is equivalent 
(3). In environmental physiological re- 
search tyramine has supplanted the use of 
NE as an indicator of the level of cold ac- 
climation (3-6). 

There is some information suggesting 
that the function of the adrenosympathetic 
nervous system is modified consequent to 
the aging process (7-11). There are two 
primary questions: (a) do older animals re- 
spond differently to equivalent doses of a 
chemical substance than younger animals; 
and (b) are there differences in response 
between various strains of a species? 
Heroux et at. (3) reported that the optimal 
dose of tyramine to elicit a maximum meta- 
bolic response in young male Sprague- 
Dawley (S-D) rats was 20 mg/kg body 
weight intraperitoneally. It is of practical 
interest to determine whether this optimal 



' To whom all correspondence should be addressed. 



dose of tyramine for young animals is opti- 
mal for older (2 years old) animals. Avail- 
able data revealed that male Sprague-Daw- 
ley rats have a median survival rate of 23 
(12), 24.5 months (13, 14), and 30 months 
(15), while male Fischer 344 (F344) rats 
have a median survival rate of 27.5 (16) to 
29 months (17). There are significant differ- 
ences in mean body weights between the 
two strains of male rats. At 80 to 90 days of 
age, male Sprague-Dawley had a mean 
weight of 325 g while F344 weighed 214 g 
(18). At 20 to 24 months, male Sprague- 
Dawley attained mean body weights of 700 
g (13) while male F344 had a mean weight of 
329 g (19). 

There is minimal information differ- 
entiating the responses of these strains to a 
stimulus, either chemical or physical. Such 
knowledge is imperative in selecting the 
appropriate strain of animal for experi- 
mental use. This paper presents the differ- 
ences in metabolic response to tyramine in 
relation to age; i.e., young and old, and 
between the two strains of rat commonly 
utilized for aging research, Sprague-Daw- 
ley and Fischer 344. 

Materials and Methods. Male Sprague- 
Dawley and Fischer 344 rats, 3 to 4 and 
20 to 24 months of age, were studied. 
The animals were individually caged and 
provided with Purina Laboratory Chow and 
water ad libitum, A 12-hr light and dark 
cycle was maintained. Oxygen uptake was 



242 

OOJ7-9727/Syil0242-05$0l. 00/0 

yrighr (<■: 1982 by the Society for Expeiimental Biology and Medicine. 
fA/s reserved. 



METABOLIC RESPONSE TO TYRAMINE IN RATS 



243 



continuously measured in a constant vol- 
ume, closed system (Volume Meter, Med 
Science, Model 160) with the rat in a sealed 
Plexiglas chamber while immersed in a 28"" 
water bath, and the expired CO2 was ab- 
sorbed by soda lime. Colonic and chamber 
temperature were measured continuously 
with thermistor probes and recorded on a 
dual-channel recorder. Before evaluating 
the animals' response to tyramine, O2 up- 
take and colonic temperature were first 
measured at 28'' for 30 min to establish a 
baseline. Each animal was then temporarily 
removed from the chamber and given in- 
traperitoneal tyramine -HCl. The animal 
was then returned to the chamber and mea- 
surements resumed for another hour. The 
resulting data were plotted and the peak re- 
sponse to tyramine which was generally 
present for several minutes was calculated. 
At least one week elapsed between each 
administration of tyramine in the same 
animal. 

The optimal dose of tyramine -HCl ad- 
ministered was determined by constructing 
a log-dose -response (LDR) curve for each 
of the four groups of rats. That is, succes- 
sively increasing log doses of tyramine - 
HCl were administered to the rats, and 
the metabolic response was plotted against 
log dose until the response curve pla- 
teaued when a higher dose of tyramine- 
HCl was administered. The optimal dose 
was defmed as the lowest dose of tyra- 
mine -HCl that produced a maximal re- 
sponse increase in O2 uptake and colonic 
temperature. The significance of the differ- 
ences between O2 uptake and colonic tem- 
perature before and after administration of 
tyramine was determined by using paired / 
tests. 

Results. The increases in O2 uptake and 
colonic temperature following tyramine in- 
jections were statistically significant from 
each basal value in all rats and at all dos- 
ages tested (p < 0.001), except in the old 
Sprague-Dawley rats when tested with the 
20 mg/kg dose. The optimal dose (20 mg/kg 
bw) for young Sprague-Dawley failed to 
elicit a significant increase in either O2 up- 
take or colonic temperature in old Sprague- 
Dawley rats. 



Young rats. Young rats respond to an in- 
jection of tyramine with increases in O2 
uptake and colonic temperature. O2 uptake 
usually peaks within 10 to 20 min following 
the injection and then gradually subsides 
and returns to basal levels in an hour or so. 
Peak response increases with increasing 
dosages until an optimal dose is reached 
and a plateau in response occurs. Elevation 
of the colonic temperature follows the rise 
of O2 uptake with a time lag of 5 to 10 min. 
The degree of elevation corresponds with 
increasing dosage and then plateaus when 
the optimal dose is reached. 

In young Sprague-Dawley (n =9), three 
dosages (S, 10, and 20 mg/kg body weight) 
were used and 20 mg/kg body weight was 
determined to be the optimal dose (Figs. 1 
and 2). A higher basal metabolic rate was 
observed in the young Sprague-Dawley 
given 20 mg compared to those given 10 mg. 
This change was probably related to the 
2-week interval between test sessions when 
these young rats grew rapidly and effect of 
age on metabolism became noticeable. 
Since in some rats this dose elevated col- 
onic temperature to 40.6'', any additional 
rise of colonic temperature was considered 
to be potentially harmful, and since in addi- 
tional studies on cold-acclimated animals 
this tyramine dose resulted in colonic tem- 
perature approaching 42'', no higher dosage 
than 20 mg/kg body weight was attempted 
in this group. 

Young F344 (n = 13) were initially tested 
with 5, 10, and 20 mg/kg body weight dos- 
ages. No obvious differences in their re- 
sponse could be discerned. Therefore, an- 
other group of young F344 (n = 13) were 
reevaluated at lower dosage levels (0.5, 1, 
and 2 mg/kg body weight). The optimal 
dose for young F344 was thus determined 
to be 2 mg/kg body weight (Figs. 1 and 2). 

Old rats. Basal O2 uptakes are lower 
in old animals, approximately 16 to 19 
mlkg~*-min"\ as compared to 21 to 27 
ml kg"' -min"* in young animals. The old 
Fischer rats have higher basal colonic tem- 
perature (approximately 0.8** higher) than 
young Fischer. However, old animals also 
respond to tyramine by increases in O2 up- 
take and colonic temperature. Most signifi- 



244 



METABOLIC RESPONSE TO TYRAMINE IN RATS 



70r 



60 - 



50 - 



40!- 



E 30 - 



20 - 



■ YOUNG SPftAGue DAWLEV , 3 monrhi ii*9 

H VOUNG FISCHER 344, 1 montM n-l3 

H OLD FISCHEf^ 344, 20 mDnthi n * |£ 

^ OLD SPftAGUE DAWLey* 24 monttu n*22 




05 1.0 20 5 

TYRAMINE (mg/kg body weight) 

Fig. 1. Increases in oxygen uptake (clear portions of the histograms) in response to intraperitoneal 
injections of tyramine. All measurements were made in an ambient environment of 28^. Error bars 
represent SEM. 



41 r 



40- 



39^ 



381" 



2 



37!- 



o 
^ 36 



35 



r^!^ 



■ YOUNG SPRAGUE QAWLEV^ 3 month* n • 9 
KyOUNS FISCHER 544 » 3 mrniffr* n-l3 
S OLD FISCHER 344 , 20 montht n- 12 
S^i-Cf SPRAGUE DAWLEY , 24 month* n-EE 



I 




05 



20 



40 



10 2.0 5 10 

TYRAMINE (mg/kg body weight) 
Fic. 2. Hlevation of colonic temperature in response to intraperitoneal ir\jections of tyramine. All 
measurements were made in an ambient environment of 28". Error bars represent SEM. 



METABOLIC RESPONSE TO TYRAMINE IN RATS 



245 



cantly, the magnitude and time course of 
the response differs strikingly from the re- 
sponses of the younger animals. The eleva- 
tions of O2 uptake and colonic temperature 
are smaller. The time course of change of 
O2 uptake is similar to that of the young, but 
the time course for the change of colonic 
temperature is altered. Peaking of colonic 
temperatures is usually delayed 50 to 60 
min. Two hours or more are required before 
basal levels are reattained. 

The optimal dose for old F344 (w = 12) 
was determined to be 5 mg/kg body weight. 
Although the increase of O2 uptake was 
similar at 2, 5, and 10 mg/kg doses, the ele- 
vation of colonic temperature apparently 
peaked at 5 mg and plateaued at 10 mg 
(Figs. 1 and 2). The optimal dose (20 mg/kg 
body weight) for young Sprague-Dawley 
did not elicit any significant response in old 
Sprague-Dawley (n = 5) (Figs. 1 and 2). A 
dose of 40 mg/kg body weight was then 
given and significant increases in O2 uptake 
and colonic temperature (P < 0.001, /f = 22) 
were obtained. No higher dosage was at- 
tempted at this point; the optimal dose for 
old Sprague-Dawley being apparently 
greater than 40 mg/kg body weight. 

Discussion. The optimal doses of tyra- 
mine for each group of rats studied (2 
mg/kg body weight for young F344, 5 mg/kg 
body weight for old F344, 20 mg/kg body 
weight for young Sprague-Dawley, and 
>40 mg/kg body weight for old Sprague- 
Dawley) represent a significant difference 
between young and old in each respective 
strain and between strains of either young 
or old animals. 

The sluggish and prolonged response in 
old animals may be explained by altered ab- 
sorption, metabolism, or excretion of drugs 
in old animals (20, 21). It is known that the 
myocardial function and blood flow to or- 
gans and tissues in old animals are reduced 
(22). There are functional alterations in 
liver and kidney (22). However, the high 
optimal doses found for old animals cannot 
be explained by these altered mechanisms 
alone, since old animals had been shown to 
be more sensitive to drugs, especially to 
norepinephrine (8), than young ones (21). 
Since tyramine acts by release of endoge- 



nous NE, this increased rather than de- 
creased dose requirement in older animals 
could be due to: 

(1) Decreased NE synthesis in old ani- 
mals. Aging impairs the uptake of tyrosine 
or dopa into neurons as well as the in- 
tracellular hydroxylation of tyrosine and 
the /3-oxidation of dopamine (7). Therefore, 
the available NE for release by tyramine is 
smaller, and more tyramine is required to pro- 
duce an apparent effect. Diminished avail- 
ability of NE can also be responsible for 
the reduced metabolic effect observed in 
old animals. 

(2) Decreased re-uptake of NE in nerve 
endings in aging animals. Since tyramine is 
taken up at the nerve ending by the same 
amine-pump mechanism, the uptake of 
tyramine can also be slowed (7). Reduced 
re-uptake of NE indicates a reduced avail- 
ability of NE for release by tyramine. 

(3) Increased monoamine oxidase (MAO) 
activity in old rats. Both NE and tyra- 
mine are substrates for MAO. Increased 
MAO activity may result in reduced avail- 
able NE and tyramine (23). 

The prolonged response or decreased 
rate of recovery in old animals represents 
an example of the often observed phenom- 
enon that the aged require more time than 
the young to reestablish homeostasis when 
displacements are induced (22). It is also of 
interest that we found a relatively lower 
basal O2 uptake but higher basal colonic 
temperature in old F344 than in young 
F344. This may suggest more efficient heat 
production, less effective heat dissipation, 
or better insulation in older F344. 

The strain-related difference observed in 
this study is of considerable significance 
since Sprague-Dawley and F344 are the 
two most commonly used strains in age- 
related studies. Mazze et aL (24) compared 
the rate of metabolism of methoxyflurane to 
inorganic fluoride and in its nephrotoxic 
effects among five rat strains: Fischer 344, 
Buffalo, Wistar, Sprague-Dawley, and 
Long -Evans. F344 rats were found to be 
more susceptible to inorganic fluoride than 
all other strains. They suggested that genet- 
ic factors were responsible for the differ- 
ence. Although some other unknown fac- 



246 



MeTABOUC B£SrCM«fE TO TYKAMKE OS BAIS 



uv% msiy be reHWttMbk for die ififlEcrewc is 
f€%tHm%€ between die F344 amd Spnigne' 
Dawley raili observed in dm study^ we aie 
imAiticd to accept die Mintuioo diat ^met- 
fc factiofi were re%pcM»aile fkir die diflEcr- 
ence. 

From our data it it apparent that gen- 
eralizations at to the pbytiological al- 
teration% induced or present in aging ani- 
mals cannot be made iigbdy. Whether or 
not the strain differences observed have a 
direct effect on aging processes remains to 
be determined. 

T1ii% research wa» utppontd in part by NationI fai- 
Mitiitc% of Health Grant NIH AG 05204413 and by the 
hvciyn Mayer Che%inui fund. 




\, TaJlman iF\ Saaveilfa iM, Axelrod i. Biotyn- 
ihe%j% and metaboli%m of endogenoui tyramine 
and it* normaJ pretence in sympathetic nerves, i 
Pharmacol KxpTher 199:216-221, 1976. 

2, Axelrod i, Gordan E. Hertiing GG, Kopin U, 
Fmter LT. (>n the mechanism of tachyphylaxis to 
tyramine in the isolated rat heart. Brit J Phar- 
macol 19:56 63. 1962. 

3, Heroux O. Pag^ H. U Blanc J. Leduc i, Gilbert 
K. Viliemaire A. Rivest P. Nonshivering ther- 
mogenesis and cold resistance in rats under severe 
cold conditions. J Appl Physiol 38:436-442, 1975. 

4, Kiang-Ulrich M. Horvath SM. Effects of cold ac- 
climation and 6-011 DA sympathectomy on cold 
tolerance of rats. J Therm Biol 4:233-237. 1979. 

5. Kiang-Ulrich M. Horvath SM. Successful cold 
acclimation following bilateral adreno-demedul- 
lation in rats. Proc Soc Hxp Biol Med 162:449- 
411. 1979. 

6. Kiang-Ulrich M. Horvath SM. Effects of cold ac- 
climation and 6-()HDA sympathectomy on cold 
tolerance of bilaterally adrenodemedullated rats. J 
Ihcrm Hiol 3:179 183. 1980. 

7. (icy Kl'. Hiircard WP. PIctscher A. Variation of 
the norepinephrine metabolism of the rat heart 
with age. (ierontologia 11:1-11. 1965. 

K. I mlkiN VV. Be/mkov VV. Bogatshaya LN. Ver- 
khiatJ«ky NS, Zamostian VP. Shevtchuk VG. 
Shichcgolcva IV. Catecholamines in the metabo- 
lism and functions regulation in aging. Geron- 
\o\om 16:129 140. 1970. 

^i liccilman IS. Ohuchi T. (Joldslcin M, Axelrod F, 




Rfib S. Dmm i. 

iiZMc3l#-3n. 
EG. McGccr HL. Ik 
iem YortL PlriiiB pS?. 
G. Cmdmi} M. < 
TiMK levels of 5s 
nts. i GcnMMl J2J92-9M. 1977. 
ir 3omn DC. KoKldoff Dl. 

m the anle m. i Gcrooiol lfc316-121. 1963. 

13. LeHcrGT.De«ttchS.liariciiidE>i.Aci^iBthe 

bod) coapoutkM. Aflier i Physiol 225:1472- 
1478. 1973. 

14. Berg BN. HaratsoB CR. Grovih. dbcase and 
agng in the rtt. i Gerontol 12:570-377. 1957. 

15. AdetaBn RC. Loss of aubplive bccIiiibsiiis dur- 
iflf ^iflf . i=ed Proc 18:1968- 1971. 1979. 

16. Hoffinian HI. Survival distnbutioQS for selected 
laboratory rtt stiaiiis and stocks. DHEW Publi- 
cation No. NIH 79-161. PP19-34. 1979. 

17. Coleman GL. Bartbold SW. OsbaMistan GW, 
Foster SJ, Jonas AM. Pathological changes during 
aging in barrier-reared Fischer 344 male rats. J 
Gerontol 32:258-278. 1977. 

18. Sines JO. Strain differences in activity, emotion- 
ality, body weight and susceptibility to stress in- 
duced stomach lesions. J Genet Psychol 
101:209-217. 1962. 

19. Chesky JA. Rockstein M. Life span characteris- 
tics in the male Fischer rat. Exp Aging Res 
2:399-407. 1976. 

20. Kato R. Vassanelli P. Frontino G. Chiesara E. 
Variation in the activity of liver microsomal 
drug-metabolizing enzymes in rats in relation to 
the age. Biochem Pharmacol 13:1037-1051. 1964. 

2 1 . Richey DP. Bender AD. Pharmacokinetic conse- 
quences of aging. Annu Rev Pharmacol Toxicol 
17:49-65. 1977. 

22. Shock NW. In: Powers JH, cd. Surgery of the 
Aged and Debilitated Patient Philadelphia. Saun- 
ders. pplO-43. 1968. 

23. Lai FM. Berkowitz B. Spector S. Influence of age 
on brain vascular and cardiovascular monoamine 
oxidase activity in the rat. Life Sci 22:2051-2056. 
1978. 

24. Mazze RI. Cousins MJ. Kosek JC. Strain differ- 
ences in metabolism and susceptibility to the neph- 
rotoxic effects of methoxyflurane in rats. J 
Pharmacol Exp Ther 184:481-488. 1973. 

Received April 5, 1982. P.S.E.B.M. 1982. Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 247-2S0 (1982) 



Reserpine Inhibits Rat Anterior Pituitary IHormone Secretion in Vitro: 
Effects on GH, TSH. and LH (41506) 

IVAN S. LOGIN,*' ALLAN M. JUDD, MICHAEL O. THORNER, and 

ROBERT M. MACLEOD 

Departments of Neurology* and Internal Medicine, University of Virginia School of Medicine, 

Charlottesville, Virginia 22908 



Abstract. Wc have extended our observations on the ability of reserpine to inhibit secre- 
tion of anterior pituitary hormones in vitro. The release of newly synthesized pH]GH and 
radioimmunoassayable TSH and LH from female rat anterior hemipituitary glands was 
measured after 5 hr incubation in pH]leucine in the presence or absence of reserpine. 
Reserpine, 9 fiM^ blocked the release of each hormone stimulated by 50 mM K* but not 
basal hormone secretion. Stimulation of ['HJGH release by 10 ^iM PGEi, or 5 mM dbcAMP 
was not affected by 5 fiM reserpine, while stimulation of TSH release by 70 nM TRH was 
significantly blunted with 5 fiM reserpine. The influence of reserpine on pituitary hormone 
secretion may deflne a pattern suggesting that the drug interferes with utilization of the 
extracellular but not the intracellular calcium mobilized during the secretory process. 



Reserpine is usually administered in vivo 
to effect biogenic amine depletion through 
blockade of aminergic reuptake mecha- 
nisms. Used in this manner, reserpine 
causes hyperprolactinemia indirectly through 
dopamine depletion and thus reduction of 
the inhibitory tone exerted on the lacto- 
troph (1). Reserpine in vitro, however, in- 
hibits prolactin secretion from rat anterior 
pituitary glands (2, 3). This unexpected 
effect was dose-related (between 0.09 and 
9 /jM) and apparently independent of the 
traditional aminergic interactions of re- 
serpine (4). The present study was de- 
signed to explore further the effects of re- 
serpine in vitro on the secretion of other 
pituitary hormones. We have investigated 
basal and stimulated secretion of newly 
synthesized ^H-labeled growth hormone 
(pH]GH) and radioimmunoassayable LH 
and TSH from female rat anterior pituitary 
glands in the presence or absence of reser- 
pine. 

Methods. Hormone secretion was evalu- 
ated by established techniques (5). Briefly, 
anterior pituitary glands from adult female 
Sprague-Dawley rats of 200-220 g (Do- 
minion Laboratories, Dublin, Va.) were 



' To whom all correspondence should be addressed. 



removed and bisected prior to 1000 hr. 
Three hemipituitaries from three different 
rats were pooled, weighed, and placed into 
an incubation flask containing 1 ml of tissue 
culture medium M-199 (M.A. Bioprod- 
ucts, Walkersville, Md.) or Earle's MEM 
(GIBCO Labs, Grand Island, N.Y.) and 10 
fiCi pHJleucine (40-50 Ci/mmole, Amer- 
sham, Arlington Heights, 111.). Each ex- 
perimental group had four flasks. The flasks 
were then incubated for 5 hr under 95% 
02-5% CO2 on a Dubnoff shaker at 37*. 

Samples of incubation medium were as- 
sayed for pH]GH, TSH, and LH. Secretion 
of newly synthesized pH]GH was mea- 
sured by polyacrylamide gel electrophore- 
sis and liquid scintillation spectrometry 
with the incorporated radioactivity ex- 
pressed relative to anterior pituitary wet 
weight as counts per minute pH]GH per 
milligram pituitary (5). Standard double 
antibody radioimmunoassay was used to 
measure TSH and LH with reagents kindly 
supplied by Dr. A. F. Parlow and the 
NIAMDD. These results were expressed as 
micrograms hormone per milligram pitu- 
itary relative to the reference preparations 
TSH RP-1 and LH RP-1. The intraassay 
coefficients of variation for TSH and LH 
were 10 and 9.8%, respectively. 

Experimental agents were added to the 



247 
0037-9727/^K V VMM -^iWi\ .^^ 

Att Tiihts Tt«entA. 



>««« 



fC£im»v« 



^KsMw^i ^fwf »4«rtvAi jE»%»M» >4flfn«0a6tt: fat: 
K^.«<r<r^v< «/ /^ tf (># 4 ^tf Xittrmi^i^r wsex*:^ 

f\fft9tii^0'A inr^'ifiyf^ ir^^t^xisn < fcsimir- 
"*^(M .^ >Mr ^f*#i . vT Vivn >:lfaie Trf iff- 

;5*fi^ H vs^fi^^t H#t w-^ .fimnlAtUin if jt- 
Y'Aj\ wf^\^'A *A^i^%v\^A ^A HiC/ X ^*»ir vw 






Tifc 



IT ":^ K T!^ 



'^Et :ti2ii: 




2: vr ^serTBoe: Fi^ ) Sestaae if 
l.'.M *r •lanmirt -nr irrl a, j ie3 «i"Mn* iff jJC 

.esn*ruirr ir «=wv -v? 

ifitt nrmiafftT m tifc im^j 

::«icnifir ttoencr^ ti icmcr/eliis- rrfjrgr ?. ^. 

mnhffK iie -.^ecrenair if iHnsr ^iniimc? 
Mimuinex imosr l Mncry if 7iin..iTL jonfi- 
iniK. inmuil ^urr ifr l.~ ?^.^ AcAMP. 
ifttt nb£ siiin laye icscszive -rffKirryfr js ^e 
iecrsnim if nnucur? Tunnme^ mbl "Ats^ 
^riiOics iicsr smamisa lUTniiiiiif abisc 
"tinii^ ioecmc. neunamsiiK. 

£.{rr:ics:Iuiar 'rwTTiim si- Trtnnrat isr i^xli 
-He SL'-'Hiiiiuiacda -eeusie if iuttiiuik^ jod 
lesirirrucHinndr^ ti ma jinuciri srcfac- 



•d^*J ' ..1 ii* » 




:« ■ . . i 




:-ir|!»»- .i-»^" ■^|..*>:3H 


Gt=. 


7^J=. ..v: LH 




inne 


'*e 


^isuisi tfTi\ tie nczimuiLin maliiiin 


. ,W.M* f.-*> Vt«'««M •*•<*'* 








LH 

,j«!ii«?i£iiiian» 




::-.r:- _ : ::•' 
y. r? : : >*' 
:> f: r : ::' 






6* r « (OC 

161 r0.ir 

1 46 r OCT- 



,f .'Ir -,-,* »♦ ^' » »/i. ,/»/tA'/ ,r'; u^**,*ti.''fifAt f t)A(A\ ukf.r. from i<» r^r^ Eich ^aiue rcprrsents the homiooe 

'. I' .■ '» '^ .*,t.p 't.*- '. }ff ,t.- .,»,-*»,/, ^ ■, . '*,* tt,4-^t, • M. 1f/f tt^, af.'.'iiip NeA!> ^>n!h<'iized "^H^OH »as measured 

'. / r .1/ • tyi,tt.:'t' p/'\ ' |A' *f.i,hf,f^ .1- -i/.'l h/ffi'l ^ pfifilUiiofi '^'.hniqucH. «hile TSH and LH were quantitated 

'. / t .^ .;ft.tf. it.'t 1 • / 

I h h\ ',itti>-ift li hf lift t* fn'h/t V. Uf'***t' 



RESERPINE DIRECTLY BLOCKS HORMONE RELEASE 



249 



TABLE H. The Effect of Reserpine on pH]GH Secretion Caused by 
PGE, OR DiBUTYRYL cAMP (dbcAMP) 



Treatment groups 



pH]GH released into the incubation medium 
(cpm/mg pituitary) 



Experiment 1 
Control 
10^ PGE, 
10 fiM PGE, + 5 fiM reserpine 

Experiment 2 
Control 

5 mAf dbcAMP 
5 mA/ dbcAMP + 5 ^iM reserpine 



222 ± 50 
734 ± 88° 
609 ±46* 

800 ± 129 
2555 ± 191° 
2114 ±346* 



Note. The incubation parameters were as described in Table 1. 

« P < 0.01 compared to the respective control. 

^ P > 0.05 compared to treatment without reserpine. 



tin secretion (7). The present findings show 
that reserpine caused a partial to complete 
blockade of the K^-mediated stimulation of 
pH]GH, TSH, and LH (Table I), and in 
previous studies the compound inhibited 
prolactin secretion (2, 3). Using other sys- 
tems reserpine blocked the release of 
[^H]DA from rat striatum (8) and nor- 
epinephrine from rabbit heart (9). These 
interactions suggest that reserpine may 
prevent effective utilization of extracellular 
calcium. 

Stimulation of GH secretion by PGEi and 
dbcAMP was not influenced by reserpine 



> 

oc 
< 



? 



20 



5 mM 
reserpine 



70 nM 
TRH 



reserpine 

A 

TRH 



Fig. 1. The effect of reserpine on TRH-stimulated 
secretion of TSH in vitro. This experiment included 
eight flasks in each group, and the ordinate represents 
/xg TSH/mg pituitary as the mean ± SE. Reserpine 
reduced the stimulatory effect of TRH on TSH release 
{P < 0.01 vs TRH alone). 



(Table II). The action of these two se- 
cretagogues, unlike K^-mediated secretion, 
is associated with intracellular calcium 
mobilization but not influx of extracellular 
calcium (10-12). 

Stimulation of TSH secretion by TRH 
may depend on both transport of extracel- 
lular and mobilization of membrane-bound 
intracellular calcium (13, 14). We found that 
during a 5-hr incubation the TRH effect was 
significantly but only partially inhibited by 
reserpine (Fig. 1). 

Hormone secretion in the presence of re- 
serpine seems to define a pattern from 
which some inferences may be drawn, al- 
though we have no direct information on 
the mechanism(s) involved. It is unlikely 
that reserpine is altering membrane prop- 
erties in a general manner to inhibit secre- 
tion since (1) the effects are selective and 
(2) hormone release can still be stimulated 
by certain secretagogues. The inhibitory 
action of reserpine is most pronounced in 
those secretory events which are strongly 
dependent on extracellular calcium, such as 
basal prolactin and K^-mediated hormone 
secretion. Reserpine is much less active 
with secretory events dependent largely on 
intracellular calcium. Studies are in prog- 
ress to evaluate the hypothesis that reser- 
pine interrupts calcium uptake in its action 
to inhibit hormone secretion. 

We appreciate the contributions of Catherine T. Har- 
cus, Suzanne B. O'Dell. Beverly Boykin, and Carlos 



250 



RESERPINE DIRECTLY BLOCKS HORMONE RELEASE 



A. Valdenegro in completing this work and the efforts 
of Courtney Ross in preparing the manuscript. Ivan S. 
Login is a recipient of Teacher Investigator Develop- 
ment Award 5 K07 NS00454 from NINCDS. Addi- 
tional support was provided by BRSA5-SO7RR05431 
(ISL), NINCHHD 5 R01-HD13197 (MOT), and 
USPHS CA 07535 (RMM). 

1. MacLeod RM, Fontham EH, Lehmeyer JE. Pro- 
lactin and growth hormone production as influ- 
enced by catecholamines and agents that affect 
brain catecholamines. Neuroendocrinology 
6:283-294. 1970. 

2. MacLeod RM. Influence of norepinephrine and 
catecholamine-depleting agents on the synthesis 
and release of prolactin and growth hormone. En- 
docrinology 85.-916-923, 1969. 

3. Login IS, MacLeod RM. The direct effect of re- 
serpine in vitro on prolactin release from rat an- 
terior pituitary glands. Brain Res 204:79-85, 
1981. 

4. Login IS, Thomer MO, MacLeod RM. Mecha- 
nisms of release of pituitary prolactin as directly 
influenced by reserpine. Trans Amer Neurol 
Assoc 105:189-191, 1980. 

5. Lamberts SWJ, MacLeod RM. Studies on the ef- 
fect of cyproheptadine on growth hormone secre- 
tion. Proc Soc Exp Biol Med 162:116-120, 1979. 

6. Douglas WW, Poisner AM. Stimulus-secretion 
coupling in a neurosecretory organ: The role of 
calcium in the release of vasopressin from the 
neurohypophysis. J Physiol 172:1-18, 1964. 

7. Thomer MO, Hackett JT, Murad F, MacLeod 
RM. Calcium rather than cyclic AMP as the 



physiological intracellular regulator of prolactin 
release. Neuroendocrinology 31:390-402. 1980. 

8. Dyck LE, Boulton AA. The effect of reserpine 
and various monoamine oxidase inhibitors on the 
uptake and release of tritiated meU-tyramine, 
para-tyramine and dopamine in rat striatal slices. 
Res Conmiun Psychol Psychiatry Behav 5:61 -78, 
1980. 

9. Misu Y, Kubo T, Nishio H. Acute partial sympa- 
thetic blockade by reserpine in the isolated rab- 
bits' hearts. Eur J Pharmacol 19:267-275, 1972. 

10. Eto S, Wood JM, Hutchins M. Fleischer N. Pitu- 
itary **Ca*' uptake and release of ACTH, GH and 
TSH: Effect of verapamil. Amer J Physiol 
226:1315-1320, 1974. 

11. Kraicer J. Spence JW. Release of GH from 
purifled somatotrophs: Use of high K" and the 
ionophore A23187 to elucidate interrelations 
among Ca*", cAMP and somatostatin. Endocri- 
nology 108:651-657, 1981. 

12. Milligan JV, Kraicer J. **Ca uptake during the in 
vitro release of hormones from the rat adeno- 
hypophysis. Endocrinology 89:766-773, 1971. 

13. Moriey JE. Neuroendocrine control of thyrotropin 
secretion. Endocrinol Rev 2:396-436, 1981. 

14. Geras E, Rebecchi MJ, Gershengoro MC. Evi- 
dence that stimulation of thyrotropin and prolac- 
tin secretion by thyrotropin-releasing hormone 
occur via different calcium-mediated mecha- 
nisms: Studies with verapamil. Endocrinology 
110:901-906, 1982. 



Received April 19, 1982. P.S.E.B.M. 1982. Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 251-257 (1982) 

6-Keto-PGFia Synthesis in Diabetic Rat Aorta: Effect of Substrate Concentration 

and Cholesterol Feeding (41507) 

H. WEY AND M. T. R. SUBBIAH^ 

Departments of Medicine and Pathology, University of Cincinnati Medical Center, Cincinnati, Ohio 45267 

Abstract. The effect of short-term feeding of a 1% cholesterol diet to normal and 
streptozotocin-induced diabetic rats on aortic 6-keto-PGFto synthesis from exogenous and 
endogenous arachidonic acid (A A) was investigated. Diabetes and cholesterol feeding (by 
themselves) resulted in a reduction in aortic synthesis of 6-keto-PGFto from both exogenous 
and endogenous arachidonic acid. An additive effect of diabetes and cholesterol feeding 
together was found for synthesis of 6-keto-PGFta from exogenous but not endogenous AA. 
Experiments in which the AA concentration was varied suggested that the inhibition of 
aortic 6-keto-PGFta synthesis by diabetes was competitive in nature. The diabetic rat was 
also found to be severely compromised in its ability to handle dietary cholesterol, as evi- 
denced by a dramatic increase in plasma total cholesterol. 



Epidemiological studies have suggested 
that patients with diabetes mellitus have an 
increased susceptibility to atherosclerosis 
(1, 2). Abnormalities of platelet function 
such as increased sensitivity to ADP, epi- 
nephrine, collagen, and arachidonic acid- 
induced aggregation have been noted (3,4). 
This is of importance since altered platelet 
function may contribute to the atheroscle- 
rotic process. 

Platelets from diabetic patients appear to 
synthesize greater amounts of the proag- 
gregatory thromboxane Ag (TXA2) (5-8). 
Aortic tissue of human diabetics (9), 
streptozotocin-induced (10, 11), and spon- 
taneously (12) diabetic rats show a reduced 
production of the antiaggregatory prostacy- 
clin (PGI2) as measured by its stable break- 
down product 6-keto-PGFia or bioassay. It 
has been suggested that the balance be- 
tween these two prostaglandin-like sub- 
stances (TXA2/PGI2) may be an important 
determinant in thrombotic disorders of dia- 
betics (11). 

Little is known about the mechanism of 
the decreased aortic PGI2 synthesis in dia- 
betes. Gerrard et aL (11) demonstrated that 
release of the substrate arachidonic acid 
from phospholipids was not impaired in the 
aorta of streptozotocin-induced diabetic 



* To whom all correspondence should be addressed. 



rats. Rather, the defect appeared to be in 
the conversion of free arachidonic acid to 
6-keto-PGF,a, although the reason for this 
defect was not determined. Furthermore, 
how diabetic rat aorta will respond to a 
stress such as cholesterol feeding, in terms 
of 6-keto-PGF,a formation, is not known. 

In this communication, we have attempt- 
ed (a) to study the kinetics of 6-keto-PGFto 
formation from exogenous arachidonic acid 
in control and diabetic rat aortas and (b) to 
assess the sensitivity of normal and diabetic 
rat aorta to the short-term feeding of a 1% 
cholesterol diet, in terms of 6-keto-PGFia 
formation from both exogenous and endog- 
enous arachidonic acid. 

Materials and Methods. Animals, Ran- 
domly selected male Sprague-Dawley rats 
(Harlan Animal Supplies, Indianapolis, 
Ind.), initially weighing between 250 and 
300 g, were made diabetic by intravenous 
(tail vein) injection of 50 mg/kg body weight 
streptozotocin (Sigma Chemical Co., St. 
Louis, Mo.) in citrate buffer (0.1 M citric 
acid and 0.145 M NaCl, pH 4.5). Controls 
were injected with an equivalent volume of 
citrate buffer. The induction of diabetes 
was confirmed 2 days later by an elevated 
plasma glucose (greater than 350 mg/dl) in 
the fed state. 

One week after injection of streptozoto- 
cin and confirmation of diabetes rats were 
randomly assigned to each of the following 



251 
0037-9727r^KVVQlS\-VI%^\.^^ 

All righXB TeMTved. 



252 



AORTIC PROSTAGLANDINS IN DIABETES 



groups: (a) control rats on control diet 
(Purina Rat Chow), (b) control rats on 1% 
cholesterol diet (Purina Rat Chow base, 
ICN Nutritional Biochemicals, Cleveland, 
Ohio), (c) diabetic rat on control diet, (d) 
diabetic rat on 1% cholesterol diet. The rats 
were maintained on these diets for 3 weeks 
and then used to study aortic 6-keto-PGFta 
synthesis. 

6'KetO'PGFia assays. [^'*C]Arachidonic 
acid (55.8 Ci/mole) was obtained from New 
England Nuclear (Boston, Mass.) and cold 
arachidonic acid from Nu Chek Prep (Ely- 
sian, Minn.). Both were stored in absolute 
ethanol at -20°. The specific activity of the 
labeled arachidonic acid was diluted with 
unlabeled arachidonic acid to obtain the de- 
sired concentrations. The ethanol was 
evaporated under N2 and the arachidonic 
acid reconstituted in 10-20 mAf NajCOg to 
the desired concentration. Fresh solutions 
in NajCOa were prepared on each day the 
assays were performed. Approximately 
0.25 /LtCi of [^^CJarachidonic acid was used 
per assay. 

The thoracic aorta was quickly dissected 
following exsanguination, cleared of con- 
nective and fatty tissue, and cut into small 
rings 3-5 mm in length. Individual rings 
were placed in 0.475 ml of phosphate- 
buffered saline (NaCl, 8 g/liter; KCl, 0.2 
g/liter; CaClg, 0.1 g/liter; MgClg 2H2O, 0.1 
g/liter; Na2 HPO4 2H2O 1.15 g/liter; 
KH2PO4, 0.2 g/liter; glucose, 1 g/liter, pH 
7.2) and preincubated for 2 min at 37° prior 
to the addition of 25 fil of various concen- 
trations of [*^C]arachidonic acid. After a 
10-min incubation the reaction was stopped 
by adding 0.5 ml 2 A/ citric acid. Time 
course experiments indicated that the aortic 
production of 6-keto-PGF,a was linear for 
at least 20 min for both control and diabetic 
aorta (Fig. 1). The prostaglandins were ex- 
tracted with 10 ml chloroform/methanol 
(2:1) and the lower organic layer was col- 
lected following the addition of 1 ml 0.9% 
NaCl. The extract was evaporated to a 
small volume under N2 and the various 
prostaglandins separated by thin-layer 
chromatography (250 /xm silica gel G plates, 
Analabs, North Haven, Conn.) using a sol- 
vent system consisting of ethyl acetate/ 




5 X) 

Incubation Time (n^mutes) 

Fig. 1. Aortic 6-kcto-PGFta synthesis (mean ± SD) 
as a function of incubation time for control (O, n =4) 
and diabetic (•, n =4) rats. 

2,5,5-trimethylpentane/acetic acid/water 
(90:50:20:100, v/v/v/v, organic layer). Prior 
to separation, pure standards for PGEj, 
PGFga, and 6-keto-PGFia (Upjohn Co., 
Kalamazoo, Mich.) were added to allow vi- 
sualization following a brief exposure to 
iodine vapor. The areas corresponding to 
PGE2, PGFaa, and 6-keto-PGFia were 
scraped into vials, 10 ml of Aquasol 2 (New 
England Nuclear) added, and quantitated 
by liquid scintillation counting. The re- 
maining areas were also scraped and 
counted to determine percentage recovery 
(always greater than 85%). Background 
counts determined from boiled aortic rings 
were subtracted prior to the calculation of 
the molar amount of product formed. 

Basal aortic G-keto-PGE^ synthesis was 
measured by specific radioimmunoassay 
(6-keto-pH]PGFia kit from New England 
Nuclear). Aortic rings were suspended in 
0.5 ml of phosphate-bufTered saline and in- 
cubated at 37° for 10 min. At the end of the 
incubation period the solution was removed 
with a pipet and frozen at -20°. The con- 
centration of 6-keto-PGFia was then mea- 
sured using radioimmunoassay. 

For both assays the amount of 6-keto- 
PGFn, formed was normalized by the sur- 
face area of the aortic ring. Surface area 
was measured after cutting the ring open 
and laying it flat. Two measurements of 
length and width were made using a Bausch 



AORTIC PROSTAGLANDINS IN DIABETES 



253 



TABLE I. Body Weight and Plasma Glucose, Triglyceride, and Cholesterol in Chow or 
Cholesterol- Fed Control and Diabetic Rats 





Body 




Concentration (ing%) 










Total 


Group 


weight 


Glucose 


Triglyceride 


cholesterol 


Control 










(n = 8) 


339 ± 11- 


186 ± 10 


27 ±6 


57 ±1 


Diabetic 










(n = 8) 


229 ± 12 


478 ± 17* 


450 ± 91* 


99 ±3* 


Control + \% 










cholesterol 










(n = 8) 


348 ±6 


177 ± 18 


66 ± 11** 


69 ±3 


Diabetic + \9c 










cholesterol 










(n=4) 


202 ± 18 


481 ± 35* 


2009 ± 388* 


3175 ± 381* 



** Results are means ± SEM. 
* /* < 0.01, as compared to the control (range test). 
** P < 0.05. 



and Lx>mb measuring magnifier equipped 
with a metric scale (one division = 0. 1 mm). 
Surface area was calculated by taking the 
product of the average length and width. 

Plasma lipid analysis. Plasma lipids were 
determined by the LRC method (14). 
Plasma glucose was determined by the glu- 
cose oxidase technique, using a Beckman 
glucose analyzer. 

Statistical analysis. Statistical compari- 
sons between means were made using the 
Newman -Keuls range test after analysis of 
variance (15). The curves drawn in Figs. 2 
and 3 were fit using weighted least squares. 
Weights were obtained from the inverse of 
the variance for each arachidonic acid con- 
centration. 

Results. The mean body weights and 
plasma levels of glucose, triglycerides, and 
cholesterol for the four groups of rats are 
presented in Table I. Diabetes caused a sig- 
nificant increase in plasma triglycerides and 
total cholesterol. Diabetic rats fed the 1% 
cholesterol diet possessed greatly elevated 
plasma triglycerides and total cholesterol. 
Control rats fed the 1% cholesterol diet ex- 
perienced only a small nonsignificant ele- 
vation in plasma total cholesterol. 

Table II shows the aortic conversion of 
exogenous [^'^CJarachidonic acid to 6- 
keto-PGFia. As previously observed by 
other investigators, diabetes resulted in a 
decreased (P < 0.05) aortic production of 



6-keto-PGF,c. Cholesterol feeding alone 
also resulted in a significant decrease in 
aortic 6-keto-PGFta production. The mean 
aortic G-keto-PCF^ production for choles- 
terol-fed diabetic rats was the lowest of 
all the groups, but only signiflcantly dif- 
ferent from the chow-fed control group. 

We also studied the effect of exogenous 
arachidonic acid concentration on the syn- 
thesis of 6-keto-PGFta by aortic rings from 
the four groups of rats. The data are pre- 
sented in Figs. 2 and 3 in the form of double 
reciprocal plots. The difference in the ca- 
pacity of control and diabetic aorta to form 
6-keto-PGF,a was found to be dependent on 
the substrate concentrations. Specifically, 

TABLE II. Conversion of 10 /iM 

["C]Arachidonic Acid to 
6-KETo-["C]PGFto IN Rat Aorta 



Group 



Aortic 6-kcto-PGFto synthesis 
(pmole/mmVIO min) 



Control (n = 8) 
Diabetic (/i = 8) 

Control + 1% 
cholesterol 
(/I = 8) 

Diabetic + \% 
cholesterol 
(/I =4) 



2.21 ±0.19- 
1.61 ±0.17* 



1.72 ±0.14* 



1.23 ±0.10* 



" Results are means ± SEM. 
* P < 0.05, as compared to the control group (range 
test). 



254 



AORTIC PROSTAGLANDINS IN DIABETES 



10 



CHOW -FED 


Km 


Vtmax 


COMTROL (O) 


12 8 


466 


DIABETIC (•) 


22 


436 




(AA) 



(MM) 



Fig. 2. Aortic 6-keto-PGF|a synthesis (pmole/mmVlO min) as a function of exogenous arachidonic 
acid concentration (AA) in control and diabetic rats. The data are presented in the form of a double 
reciprocal plot. The apparent maximum velocity (pmole/mmVlO min) and K^ (jiAf) values were calcu- 
lated from the regression lines. The number of rats are given in parentheses and the vertical bars 
represent the SD. 



differences were more evident at lower 
substrate concentrations. The calculated 
apparent maximum velocity and apparent 
K,ft values are given in Figs. 2 and 3. A 
useful comparison of the four groups can be 
obtained from these values. Diabetes re- 
sulted in an increased apparent K^ with no 
change in the apparent maximum velocity, 
as compared to the chow-fed control group. 
The apparent K^ was also increased for the 
cholesterol-fed control group and even 
more so for the cholesterol-fed diabetic 
group. Cholesterol-feeding also resulted in 
a slightly greater apparent maximum veloc- 
ity for both diabetic and control rats. 

Basal aortic synthesis of 6-keto-PGFia 
from endogenous arachidonic acid was also 
measured and found to follow a similar 
trend. Both cholesterol feeding and diabe- 
tes resulted in lower mean values as com- 
pared to the chow-fed group (Table III). 
However, the cholesterol-fed diabetic 
group did not possess the lowest mean 
value. 



Discussion. The results of the present 
study confirm previous observations (10- 
13) demonstrating that the diabetic state 
decreases the formation of 6-keto-PGFia 
in rat aorta. The results further demon- 
strate that short-term cholesterol feeding 
(3 weeks) also results in a reduced aortic 
synthesis of 6-keto-PGFi„ in the adult male 
rat. Our studies using exogenous [*^C]- 
arachidonic acid suggested that feeding 
cholesterol to diabetic rats may further alter 
aortic 6-keto-PGFic synthesis. The mag- 
nitude of the change in aortic 6-keto-PGF,a 
synthesis at 10 iiM exogenous arachidonic 
acid and the apparent K^ for the cho- 
lesterol-fed diabetic group was roughly 
equal to the sum of the separate effects of 
cholesterol feeding and diabetes. Thus, the 
reduction in aortic 6-keto-PGFi„ synthesis 
found in the cholesterol-fed diabetic group 
may simply be an additive effect of choles- 
terol feeding and diabetes. An interesting 
fmding was that diabetes appeared to result 
mainly in an increased apparent K„t with 



AORTIC PROSTAGLANDINS IN DIABETES 



255 







1-fe CHOLEST- 
EROL - FED 










Km 


Vm^x 






CONTROL (O) 


210 


5 40 




10 


DIABETIC (•) 


33 9 


4 94 












>^ ( 


i 

o 

5 
|05 




^X 


J 

(4) 


/ 


/^ C4) 




>y 


A^ 


^ 










(4) 






1 



00 



005 



01 



Fig. 3. Aortic 6-keto-PGFta synthesis (pmole/mmVlO min) as a function of exogenous arachidonic 
acid concentration (AA) in control and diabetic rats fed 1% cholesterol. The data are presented in the 
form of a double reciprocal plot. The apparent maximum velocity (pmoIe/mmVlO min) and K^ (jiM) 
values were calculated from the regression lines. The number of rats are given in parentheses and the 
vertical bars represent the SD. 



only minimal effect on the apparent maxi- 
mum velocity. This resembles a competi- 
tive type of inhibition of aortic 6-keto- 
PGFitt synthesis. Using endothelial cell 
cultures, Spector, et al. (16) found that 
linoleic and to some degree oleic acid in- 
hibited the conversion of arachidonic acid 
to 6-keto-PGFto and that this inhibition ap- 
peared to be competitive in nature. Fur- 
thermore, increasing the dietary intake of 
linoleate by feeding com oil to rabbits has 
been shown to result in a decreased aortic 
prostacyclin-producing capacity (17). Thus, 
it might be speculated that alterations in the 
levels of certain free or esterified fatty acids 
may be involved in the diabetes-induced 
reduction in aortic 6-keto-PGFia synthesis. 
Faas and Carter (18) found that liver micro- 
somal A6 desaturase activity was de- 
pressed in diabetic rats. This depression 
was presumably the cause of the elevated 
levels of linoleate found in the liver micro- 
somes. Serum-free fatty acids are also 
greatly increased in diabetic rats (19). 



These observations suggest that tissue 
linoleate would be elevated in diabetic rats. 
Dietary cholesterol has also been found 
to alter the composition of rat liver phos- 
pholipids (20). The result was an elevation 
in the percentage composition of linoleate 
and 8,11,14-eicosatrienoate of rat liver 

TABLE III. Basal Aortic 6-KETo-PGFta 

Synthesis from Endogenous 

Arachidonic Acid 

Aortic 6-keto-PGFio synthesis 
Group (pmole/mmVlO min) 



Control (n = 4) 
Diabetic {n = 4) 

Control + 1% 
cholesterol 
in =4) 

Diabetic + 1% 
cholesterol 
(n =4) 



4.07 ± 0.30« 
2.42 ± 0.43* 



3.31 ±0.21 



2.58 ± 0.44* 



<* Results are means ± SEM. 

* P < 0.05. as compared to the control group (range 

test). 



256 



AORTIC PROSTAGLANDINS IN DIABETES 



phospholipids following the cholesterol 
supplementation of a control diet contain- 
ing cottonseed oil. The mechanism of this 
alteration remains to be determined. It 
would be of interest to know if cholesterol 
feeding and diabetes alter aortic 6-keto- 
PGFto synthesis by the same mechanism. 
We found that the apparent K^ value for 
aortic synthesis of 6-keto-PGF,e, was in- 
creased for the chow-fed diabetic, cho- 
lesterol-fed control, and cholesterol-fed 
diabetic group. Only cholesterol feeding, 
however, resulted in an elevation in the ap- 
parent maximum velocity. This result 
suggests the possibility that cholesterol 
feeding may alter aortic 6-keto-PGF,„ syn- 
thesis by a mechanism different from dia- 
betes. Further research will be necessary to 
resolve the mechanisms of dietary choles- 
terol and diabetes-induced alterations in 
aortic 6-keto-PGFia synthesis. 

The results of measurements of basal 
aortic 6-keto-PGFi„ synthesis were similar 
to those using [^''CJarachidonic acid except 
that we found no additional effect of feeding 
cholesterol to diabetic rats. The reason for 
this difference is unknown. The work of 
Needleman et al. (21) suggests that the en- 
zymes involved in the synthesis of PGI2 and 
PGE2 by the perfused kidney may exist in 
multiple compartments within the cell. 
Thus, to some extent aortic PGI2 synthesis 
from endogenous arachidonic acid could 
involve different enzymes than synthesis 
from exogenous arachidonic acid. It is pos- 
sible then that measurements of aortic 6- 
keto-PGFia synthesis from exogenous and 
endogenous arachidonic acid could yield 
differing results. It should be noted that 
aortic synthesis of 6-keto-PGFi„ from en- 
dogenous arachidonic acid (molar amount) 
was greater than that from 10 fiM exoge- 
nous arachidonic acid (compare Tables II 
and III). Karpen et al. (22) have also pub- 
lished data demonstrating a greater molar 
production of 6-keto-PGF,„ by rat aortic 
rings from endogenous arachidonic acid 
versus 17.6 fiM exogenous arachidonic acid 
after a 90-min incubation at 37°. In contrast 
the addition of exogenous arachidonic acid 
to bovine aortic segments and cultured en- 
dothelial cells greatly enhances PGL syn- 



thesis over the basal level (23). The reason 
for the difference between these tissues in 
their 6-keto-PGFta synthesis from exoge- 
nous versus endogenous arachidonic acid is 
not known, but may be related to differ- 
ences in the uptake and utilization of exog- 
enous arachidonic acid, the basal capacity 
for PGI2 synthesis, or both. Further study 
will be required to answer the first possibil- 
ity. In regard to the latter possibility, it has 
been shown that PGI2 production by rat 
aorta was quite high as compared to other 
species such as rabbit and guinea pig (24). 
Sinzinger et al, (25) have suggested that the 
susceptability of various species to athero- 
sclerosis may be due to inborn differences 
in PGI2 formation. The significance of the 
difference in our findings when using exog- 
enous versus endogenous arachidonic acid 
awaits further investigation. 



1. Robertson WB, Strong JP. Atherosclerosis in per- 
sons with hypertension and diabetes mellitus. Lab 
Invest 18:538-551, 1968. 

2. Keen H. Glucose intolerance, diabetes mellitus 
and atherosclerosis; prospects for prevention. 
Postgrad Med J 52:445-451, 1976. 

3. Sagel J, Colwell JA, Crook L, Laimins M. In- 
creased platelet aggregation in early diabetes mel- 
litus. Ann Intern Med 82:733-738, 1975. 

4. Colwell JA, Halushka PV, Sarji K, Levine J, 
Sagel J, Nair RMG. Altered platelet function in 
diabetes mellitus. Diabetes 25 (Suppl 2):826-831, 
1976. 

5. Ziboh VA, Maruta H, Lord J, Cagle WD, Lucky 
W. Increased biosynthesis of thromboxane A^ by 
diabetic platelets. Eur J Clin Invest 9:223-228, 
1979. 

6. Butkus A, Skrinska VA, Schumacher OP. 
Thromboxane production and platelet aggregation 
in diabetic subjects with clinical complications. 
Thromb Res 19:211-223, 1980. 

7. Lagarde M, Burtin M, Berciaud P, Blanc M, Vel- 
ardo B, Dechavanne M. Increase of platelet 
thromboxane A2 formation and its plasmatic 
half-life in diabetes mellitus. Thromb Res 
19:823-830, 1980. 

8. Halushka PV, Rogers RC, Loadhold CB, Colwell 
J A. Increased platelet thromboxane synthesis in 
diabetes mellitus. J Lab Clin Med 97:87-%, 1981. 

9. Johnson M, Harrison HE, Raferty AT, Elder JB. 
Vascular prostacyclin may be reduced in diabetes 
in man. Lancet 1:325-326, 1979. 

10. Harrison HE, Reece AH, Johnson M. Decreased 



AORTIC PROSTAGLANDINS IN DIABETES 



257 



vascular prostacyclin in experimental diabetes. 
Life Sci 23:351-356, 1978. 

11. Gerrard JM, Stuart MJ, Rao GHR, Stcffes MW, 
Mauer SM, Brown DM, White JG. Alteration in 
the balance of prostaglandin and thromboxane 
synthesis in diabetic rats. J Lab Clin Med 
95:950-958, 1980. 

12. Subbiah MTR, Deitemeyer D. Altered synthesis 
of prostaglandins in platelet and aorta from spon- 
taneously diabetic Wistar rats. Biochem Med 
23:231-235, 1980. 

13. Rogers SP, Larkins RG. Production of 6-oxo- 
prostaglandin Fiq by rat aorta. Influence of dia- 
betes, insulin treatment, and caloric deprivation. 
Diabetes 30:935-939, 1981. 

14. LRC Manual of Labs Operation, Vol 1, Lipid and 
Lipoprotein Analysis. NHLI-DHEW publication 
No. 75-028, U.S. Govt Printing Office, 
Washington DC, 1979. 

15. Snedecor GW, Cochran WG. Statistical Methods, 
6th ed. Iowa State Univ Press, Ames, p271, 1967. 

16. Spector AA, Hoak JC, Fry GL, Denning GM, 
Stoll LL, Smith JB. Effect of fatty acid modiflca- 
tion on prostacyclin production by cultured 
human endothelial cells. J Clin Invest 65:1003- 
1012, 1980. 

17. Galli C, Agradi E, Petroni A, Tremoli E. Differen- 
tial effects of dietary fatty acids on the accumula- 
tion of arachidonic acid and its metabolic conver- 
sion through the cyclooxygenase and lipooxy- 
genase in platelets and vascular tissue. Lipids 
16:165-172, 1981. 

18. Faas FH, Carter WH. Altered fatty acid desatura- 
tion and microsomal fatty acid composition in the 
streptozotocin diabetic rat. Lipids 15:953 -%1, 
1980. 



19. Paulson DJ, Crass MF. Myocardial triacylglycerol 
fatty acid composition in diabetes mellitus. Life 
Sci 27:2237-2243, 1980. 

20. Morin RJ, Bemick S, Mead JF, Alfin-Slater RB. 
The influence of exogenous cholesterol on hepatic 
lipid composition of the rat. J Lipid Res 
3:432-438, 1%2. 

21. Needleman P, Wyche A, Bronson SD, Holmberg 
S, Morrison AR. Specific regulation of peptide- 
induced renal prostaglandin synthesis. J Biol 
Chem 254:9772-9777, 1979. 

22. Karpen CW, Pritchard KA, Merola AJ, Pan- 
ganamala RV. Alterations of the prostacyclin- 
thromboxane ratio in streptozotocin induced 
diabetic rats. Prostaglandins Leukotrienes Med 
8:93-103, 1982. 

23. Goldsmith JC, Jafvert CT, Lollar P, Owen WG, 
Hoak JC. Prostacyclin release from cultured and 
in vivo bovine vascular endothelium. Studies with 
thrombin, arachidonic acid, and ionophore 
A23187. Lab Invest 45:191-197, 1981. 

24. Tschopp TB, Baumgartner HR. Platelet adhesion 
and mural platelet thrombus formation on aortic 
subendothelium of rats, rabbits, and guinea pigs 
correlate negatively with the vascular PGI, pro- 
duction. J Lab Clin Med 98:402-411, 1981. 

25. Sinzinger H, Clopath P, Silberbauer K, Winter M. 
Is the variation in the susceptability of various 
species to atherosclerosis due to inborn differ- 
ences in prostacyclin (PGI,) formation? Ex- 
perientia 36:321-323, 1980. 



Received July 1, 1981. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOOY AND MEDICINE 171, 258-265 (1982) 



Effect of Hypertonic Sodium Chloride on Polyribosomes and Protein Synthesis of 

Kidneys of Rats^ (41508) 

CHALLAKONDA N. MURTY, BEATRICE OLIVEROS, and 
HERSCHEL SIDRANSKY^ 

Department of Pathology, The George Washington University Medical Center, Washington, D.C. 20037 

Abstract. The effects of a single intraperitoneal administration of hypertonic NaCI (5.3 ml 
oil. 55% NaCl/100 g body wt) on polyribosomes and in vitro protein synthesis of kidneys of 
rats were investigated. The results revealed that there was marked disaggregation of kidney 
polyribosomes of rat (fasted or nonfasted) that received a single dose of hypertonic NaCl 
solution 30 min prior to sacrifice in comparison to control rats that received isotonic NaCl 
solution. In vitro [*^C]leucine incorporation into protein was significantly decreased using 
either postmitochondrial supernatant or microsomes from kidneys of hypertonic NaCl- 
treated rats compared with control rats. Administration of a single dose of hypertonic NaCl 
to rats produced a significant increase in the activity of renal RNase in comparison to that of 
the control group. Assay of renal acid phosphatase in the particulate and soluble fractions 
revealed no changes in the activities in the control and experimental rats. 



Earlier studies from our laboratory have 
reported that the administration orally or 
intraperitoneally of hypertonic solutions of 
sodium chloride, potassium chloride, mag- 
nesium chloride, sodium sulfate, or sucrose 
to mice rapidly induced disaggregation of 
polyribosomes and decreased in vitro and in 
vivo incorporation of [^'*C]leucine into pro- 
teins of livers (1). In further studies, it has 
been demonstrated that ribosomes isolated 
from the livers of mice treated with a 
hypertonic (4%) NaCl solution were less 
active in initiation of polypeptide synthesis 
than those isolated from control animals re- 
ceiving a 0.85% NaCl solution (2). Initiation 
factors isolated from the livers of control 
and experimental animals were equally ac- 
tive, suggesting that the ribosomes alone 
were affected by the hypertonicity (2). The 
deleterious effects of hypertonic solutions 
on hepatic polyribosomes and protein syn- 
thesis were shown to be mediated by a rise 
in the osmotic pressure of portal venous 
blood plasma (3). 



' This investigation was supported by USPHS Re- 
search Grants CA 26557 from the National Cancer In- 
stitute and AM 27339 from the National Institute of 
Arthritis, Metabolism and Digestive Diseases. 

- To whom all correspondence should be addressed. 



The present investigation was designed 
to determine whether polyribosomes in or- 
gans other than liver will respond to 
changes in osmotic pressure of blood in- 
duced by administering hypertonic solu- 
tions. In this study, the effects of hyper- 
tonic NaCl injected intraperitoneally on 
polyribosomes and in vitro protein synthe- 
sis of kidneys of rats were investigated. The 
kidney was selected since it plays a vital 
role in the purification of blood and ex- 
cretes excess salts derived from the blood. 
Furthermore, since rat serum has been re- 
ported normally to contain high levels of 
ribonuclease activity (4), it was decided to 
determine the ribonuclease activities of 
serum and kidney after administering 
hypertonic NaCl in consideration that ele- 
vations may occur which could influence 
the status of kidney polyribosomes and the 
rate of protein synthesis. The results of this 
communication indicate that the adminis- 
tration of a single dose of hypertonic NaCl 
to rats caused disaggregation of polyribo- 
somes and inhibition of in vitro protein 
synthesis of kidneys along with an in- 
creased activity of renal ribonuclease. 

Materials and Methods. Animals. 
Female rats of the Sprague-Dawley strain 
(Sprague-Dawley, Madison, Wisconsin), 
1 1 weeks old, and weighing 170-220 g, were 



258 
?7/8yn0258-08$0L 00/0 
? /9g2 by the Society for Experimental Bio\ogy and Medicine. 



HYPERTONIC NaCI AND RENAL PROTEIN SYNTHESIS 



259 



used. The rats were maintained in a 
temperature-controlled room with alter- 
nating 12-hr cycles of light and dark. In 
studies with fasted rats, the animals were 
fasted overnight but had free access to 
water. The following morning, fasted and 
nonfasted rats were divided into groups, 
each of which contained two to four ani- 
mals. Rats received intraperitoneally either 
0.85 or 7.55% NaCl solution (45 or 400 mg/ 
100 g of body wt) 30 min before killing. 

Chemicals, L-[*^C]Leucine (10 mCi/ 
mmole) and poly-[5-^H]uridylic acid (17-61 
Ci/mmole) were obtained from Amersham 
Corporation, Arlington Heights, Illinois. 
Ribonuclease A, purified from bovine pan- 
creas, was purchased from Worthington 
Life Sciences Division, Freehold, New 
Jersey. 

Sucrose density gradient analysis of 
polyribosomes. Rats were killed by de- 
capitation and bleeding. Kidneys were re- 
moved, dissected free of perirenal fat, and 
washed three times in ice-cold 0.25 M su- 
crose solution containing TKM buffer (0.05 
M Tris-HCl, pH 7.5; 0.025 M KCl; and 
0.01 M MgCla). The kidneys were weighed, 
minced with scissors in 4 vol of ice-cold 
0.25 M sucrose in TKM buffer, and 
homogenized with a Potter- Elvehjem 
homogenizer with a motor-driven Teflon 
pestle (clearance, 0.010 in.) and four up- 
and-down strokes at 1325 rpm. Post- 
mitochondrial supernatant (PMS) was pre- 
pared by centrifuging the homogenate using 
a SS34 rotor in a Sorvall Model SS-3 cen- 
trifuge for 10 min at SCXX) rpm. A portion 
(0.4 ml containing 26 Ajeo nm units) of the 
PMS was layered on top of a linear 12-ml 
sucrose gradient (0.3 to 1.1 A/ sucrose con- 
taining TKM) in a cellulose nitrate tube, 
centrifuged using a Spinco SW 41 rotor in a 
Spinco Model L3-40 ultracentrifuge at 
38,000 rpm for 1 hr at 4° and the polyribo- 
somes were analyzed using a Gilford Model 
2400S UV spectrophotometer as described 
earlier (3). The degree of polyribosomal 
aggregation of kidneys in the control and 
experimental groups was evaluated from 
the sucrose density gradients by calculating 
the relative distribution of monomer-dimers 
in relation to total ribosomes. 



In vitro [^^C^eucine incorporation into 
proteins. In vitro incorporation of [^^C]- 
leucine into proteins using PMS or mi- 
crosomes of kidneys was performed by 
the method of NichoUs et al, (5). Micro- 
somes were prepared by centrifuging PMS 
for 1 hr at 40,000 rpm in a Spinco Model 
L3-40 ultracentrifuge. Microsomal pellets 
were resuspended in 0.25 M sucrose con- 
taining TKM buffer. 

PMS (0.5 mg protein) or microsomes 
(0.25 mg protein) were incubated with 0.5 
ml of standard incubation mixture in a total 
volume of 1.0 ml for 1 hr at 37*. The incu- 
bation mixture for each test tube contained 
ATP (2 /Ltmole), phosphoenol pyruvate (10 
/imole), pyruvate kinase (40 fig protein), 
sucrose (125 fimole), Tris-HCl, pH 7.5 (25 
/Ltmole), K^ (12.5 /itmole), Mg^^ (5 /itmole) 
and [*^C]leucine (0.5 fiCi). Cell sap (5 mg 
protein) from livers of control rats was used 
in the reaction mixture when microsomes 
were incubated. The cell sap was prepared 
by homogenizing 1 part liver and 2.5 parts 
0.25 M sucrose containing TKM buffer and 
then centrifuging at 8000 rpm for 10 min. 
The supernatant was then centrifuged at 
40,000 rpm for 2 hr. The resultant superna- 
tant (cell sap) was used for in vitro incorpo- 
ration of [^^C]leucine into protein using kid- 
ney microsomes. At the end of the incuba- 
tion period, 1 ml of 10% trichloroacetic acid 
containing 20 mg of celite was added to 
each tube and processed for the measure- 
ment of radioactivity into proteins as de- 
scribed earlier (3). 

Determination of RNase activity. The 
RNase activity in PMS of kidneys of con- 
trol and experimental rats was determined 
by the method of Liu and Matrisan (4). The 
200-/Ltl final volume of reaction mixture 
contained 0.1 M Tris-HCl, pH 7.4, 5 mAf 
CaCIa, 0.3 mAf dithiothreitol, 0.1 fiC\ 
pH]poly(U) containing 180,000 cpm and 
10-50 fi\ of PMS containing 0.5 to 2.0 fi% 
protein. At the end of 15 min incubation at 
37°, the reaction was stopped by the addi- 
tion of 0.6 ml of cold 95% ethanol contain- 
ing 10 mA/ Mg-acetate and 0.06 mg yeast 
RNA in 0.58 M Na-acetate. The samples 
were kept at -20° for 1 -2 hr and then cen- 
trifuged at 10,000 rpm for 15 min. Aliquots 



260 



HYPERTONIC NaCI AND RENAL PROTEIN SYNTHESIS 



of the supematants were counted for the 
radioactivity in the enzyme hydrolyzed, 
ethanol-soluble nucleotides. The RNase 
activity (pH 7.4, 7.6, and 9.5) in the serums 
of control and experimental animals was 
determined according to the method of 
Rahman (6). 

Determination of acid phosphatase ac- 
tivity. The acid phosphatase activity in the 
particulate and soluble fractions of kidneys 
of control and experimental rats was de- 
termined by the method of Tani and Ogata 
(7). The 3.0-ml total volume of reaction 
mixture contained 200 fimole of 0.2 M 
Tris-maleate buffer, pH 6.0, 10 fimolt of 
0.02 M pyridoxine, 50 fimole of 0.1 M p- 
nitrophenylphosphate, and 1.2-2.4 mg 
protein of the soluble or particulate frac- 
tion. The reaction mixture was incubated 
for 30 min at 37**. The reaction was stopped 
by addition of 3 ml of 50% trichloroacetic 
acid. Aliquots of 0.1 ml were diluted to 2.0 
ml with distilled water, then added 2.0 ml of 
saturated sodium carbonate andp-nitrophe- 
nol formed was determined by reading the 
samples at optical density of 430 nm. 

Chemical determinations. Protein was 
determined by the method of Lowry et al, 
(8) and RNA was assayed by the method of 
Fleck and Munro (9). 

Results. In earlier studies, isolation of 
polyribosomes from rat kidney, using pro- 
cedures that were successful with liver, 
yielded only small polyribosomal aggre- 
gates (5, 10). The difficulty in obtaining in- 
tact polyribosomes from kidney has been 
attributed to a higher level of RNase activ- 
ity in kidney than in other tissues, such as 
liver, brain, spleen, etc. However, Zomzely 
et al. (11) have reported difficulty in the 
preparation of intact polyribosomes from 
the brain tissue, similar to that with kidney, 
even though brain contains relatively low 
levels of RNase activity (lower than in 
liver). These authors overcame this diffi- 
culty during the isolation of brain polyribo- 
somes by employing high concentrations of 
Mg-'*' (10- 12 mA/) (higher than used in the 
preparation of liver polyribosomes) and by 
the omission of detergent treatment. These 
concentrations of Mg-^ maintained the in- 
tegrity of the polyribosomes. Therefore, in 



the present study, a 10 mM Mg^^ concen- 
tration was used during the preparation and 
sucrose gradient analysis of polyribosomes 
from kidney postmitochondrial supema- 
tants. Furthermore, treatment of kidney 
postmitochondrial supernatant with deoxy- 
cholate, prior to layering on sucrose gra- 
dients to examine polyribosomes as in the 
case with the liver, was omitted. This 
omission of detergent treatment of renal 
postmitochondrial supernatant should not 
influence the yield of polyribosomes since, 
unlike in the liver, the great majority of 
polyribosomes in kidney of adult rats are 
free polyribosomes (12). 

Since rat kidney has been shown to con- 
tain a high level of RNase activity, we, in 
the initial experiments, examined the influ- 
ence of adding one of two RNase inhibitors 
(liver postmicrosomal supernatant and 
polyvinyl sulfate) during the isolation and 
subsequent analysis on the status of kidney 
polyribosomes. This was conducted to de- 
termine whether inclusion of either of these 
two RNase inhibitors would result in ob- 
taining better polyribosomal profile, prepa- 
rations with heavier aggregates. The pro- 
files of kidney polyribosomes prepared in 
the presence of either one of the two RNase 
inhibitors were similar to those prepared in 
their absence (unpublished data). There- 
fore, in subsequent studies these RNase in- 
hibitors were not included during the prepa- 
ration of kidney polyribosomes. Figure 1 
illustrates the profiles of polyribosomes 
prepared from kidneys of control rats under 
our experimental conditions and reveals the 
large polyribosomal aggregates with very 
few monomers and dimers. The polyribo- 
somal patterns of the control rats appear 
to be similar to those reported by others 
(4, 10, 12). 

The effects of administering a hypertonic 
NaCl (7.55%) solution intraperitoneally on 
kidney polyribosomes of fasted or non- 
fasted rats are presented in Fig. 1. Su- 
crose density gradient patterns of kidney 
polyribosomes 30 min after the administra- 
tion of hypertonic NaCl revealed marked 
disaggregation of polyribosomes (an in- 
crease in dimers and monomers in relation 
to total ribosomes) in comparison to those 



HYPERTONIC N«a AND RENAL PROTEIN SYNTHESIS 



261 



»A»TtO r»&%>«Ct 




EFFLUfMT VOLUME (ml) 

Fig. 1. Sucrose density gradient patterns of postmitochondrial supernatant* of kidney polyribo- 
somes of rats that received 0.85 or 7.55% NaCI intniperitoneally 30 min before being killed. 



of control rats that received isotonic NaCI 
(0.8S% solution). Hypertonic NaCI caused 
marked disaggregation of polyribosomes of 
kidneys of both fasted and nonfasted rats. 
Measurements of the areas under the 
polyribosomes and under the monomers 
plus dimers of the sucrose density gradient 
patterns were made in several experiments 
and the results are summarized in Table I. 
The data indicate that the administration of 
hypertonic NaCI to fasted and nonfasted 
animals resulted in a significant increase in 
monomer plus dimers along with a con- 
comitant decrease in heavier polyribo- 
somes. 
In vitro ['^Cjleucine incorporation into 



protein using postmitochondrial superna- 
tants or microsomes of kidneys of control 
and experimental rats was also studied and 
the results are summarized in Table I. Both 
the postmitochondrial supernatants and the 
microsomes from kidneys of hypertonic 
NaCI-treated rats exhibited significant de- 
creases in in vitro ('^Cjleucine incorpora- 
tion into proteins compared to values de- 
rived using preparations from control rats. 
Generally, the disaggregation of the renal 
polyribosomes correlated with the decrease 
in in vitro protein synthesizing ability in 
both fasted and nonfasted rats receiving 
hypertonic NaCI. 
The administration of hypertonic NaCI 



262 



HYPERTONIC NaCl AND RENAL PROTEIN SYNTHESIS 



TABLE I. 



Effect of Administering Hypertonic NaCl on Weight, Polyribosomes, and 
IN Vitro Protein Synthesis of Kidneys 





Weight 

(mg/lOOg 

body weight) 


Status of polyribosomes 

(Monomer - dimers/total 

ribosomes x 100) 


[*^C]Leucine incoiporation into 
protein using: 


Group" 


Postmitochondrial 

supernatant Microsomes 
(cpm/mg RNA) (cpm/mg RNA) 


Fasted rats 
0.85% NaCl 
7.55% NaCl 
% change 

Nonfasted rats 
0.85% NaCl 
7.55% NaCl 
% change 


871 ±23(17)* 
763 ± 19* (14) 
-12.2 ± l.n*-* 

800 ± 21 (14) 
818 ±27 (14) 
+ 1.72 ±3.17*" 


31.26 ±4.63 (5) 
46.80 ± 2.75* (5) 
+60.6 ± 20.5 

32.83 ± 2.47 (6) 
52.50 ± 0.66* (6) 
+63.9 ±11.1* 


1479 ±165 (11) 1878 ±24.4 (8) 
457 ±6.7* (11) 685 ±75* (8) 
-65.2 ±5.7* -58.1 ±7.0* 

1614 ±318 (6) 3971 ±1005(3) 
439 ± 122 (6) 2017 ± 420 (3) 
-71.5 ±6.7* -45.5 ±9.1* 



" Rats were fasted overnight or were nonfasted and then received 0.85 or 7.55% NaCl solutions (5.3 ml/ 100 g 
body weight) intraperitoneally 30 min before killing. In each experiment, kidneys of three or four rats of each 
group were pooled. Weights are means of the sum of both kidneys. 

* Number of experiments in parentheses. Mean ± SEM. 
'' Mean ± SEM of differences for each experiment. 

* P < 0.05. 



did not alter the body weights of both the 
fasted and nonfasted rats. The mean body 
weights of the fasted rats were 156 ± 4.8 g 
(SEM) for the isotonic NaCl group and 156 
± 4.7 g for the hypertonic NaCl group, and 
of the nonfasted animals were 189 ± 4.7 g 
for the control group and 190 ± 4.6 g for the 
experimental group. However, the admin- 
istration of hypertonic NaCl 30 min before 
killing caused a small, statistically signifi- 
cant decrease (-12%) in kidney weights in 
the fasted rats (Table I). On the other hand, 
nonfasted rats receiving hypertonic NaCI 
had a small insignificant increase (+2%) in 
kidney weights (Table I). 

RNase activities of kidneys of control 
and hypertonic NaCl-treated rats that were 
fasted overnight were measured in six ex- 
periments. The results are summarized in 
Table II and reveal that the levels of RNase 
activity of the postmitochondrial superna- 
tants were significantly higher in the kid- 
neys of hypertonic NaCl-treated animals 
than the levels in the kidneys of the 0.85% 
NaCl-treated rats. 

It has been reported earlier that rat serum 
normally contains a relatively high level 
(higher than in serum of mice) of RNase 
activity (4). Therefore, it was of special 
interest to determine whether the elevated 



levels of renal RNase activity in the ex- 
perimental rats may be attributed to 
changes in the circulating serum RNase in 
rats treated with hypertonic NaCl. RNase 
activities at pH 5.4, 7.6, and 9.5 were mea- 
sured in the sera of control and hypertonic 
NaCl-treated rats. The results of these ex- 
periments are summarized in Table III and 
reveal no appreciable changes in the ac- 
tivities of RNase in sera of the control and 
experimental groups. 

Since the serum RNase activities were 
essentially unaltered after the administra- 



TABLE 11. RNase Activities of the 

Postmitochondrial Supernatants of Kidneys 

OF Rats That Received Isotonic or 

Hypertonic NaCI 





Renal RNase activities 


Group" 


Units/mg protein Percentage change 


Q.S5% NaCI 
7.55*^ NaCI 


14.6 ± 3.4 (6)* 

34.0 ± 7.2* (6) + 152.9 ± 42.4'-** 



" Overnight fasted rats received 0.85% NaQ or 7.55% NaCI 
solution intraperitoneally 30 min before killing. In each ex- 
periment, kidneys of three or four rats in each group were 
pooled. 

** Number of experiments in parentheses. Mean ± SEM. 

• Mean ± SEM of differences for each experiment. 
* r < 0.05. 

**P < 0.02. 



HYPERTONIC N«CI AND RENAL PROTEIN SYNTHESIS 



263 



TABLE III. RNase Activity in the Sera 


OF 


Rats Treated with Isotonic or Hypertonic NaCl 


Serum RNase activities 
(i4 too units/ml) 


Group* pH 5.4 




pH7.6 


pH9.5 


0.85% NaCl 31.97 ±4.2 (8)* 
7.55% NaCl 31.38 ±2.2 (8) 




36.08 ±2.7 (13) 
38.35 ± 1.9(13) 


38.23 ± 6.4 (8) 
37.32 ±6.1 (8) 



' Overnight-fosted rats received 0.85% NaCl or 7.55% NaCl intraperitoneally 30 min before killing. Blood 
samples were collected by heart puncture under ether anesthesia and serum samples were separated 30 min later 
by centrifiigation. 

* Number of animals in parentheses. Mean ± SEM. 



tion of hypertonic NaCl, we then consid- 
ered the possibility that the hypertonic 
NaCl induced cellular injury within the kid- 
neys by altering the osmolarity of the cir- 
culating blood. Such injury could affect the 
lysosomes within the renal epithelial cells 
and lead to an elevation in free RNase ac- 
tivity. To determine whether lysosomal re- 
lease of enzymes may be responsible for the 
elevated levels of RNase activity in the kid- 
neys of hypertonic NaCl-treated rats, we 
assayed for the activity of an important 
lysosomal enzyme, acid phosphatase, in the 
particulate and supernatant fractions of 
kidneys of control and experimental rats in 
three experiments. The acid phosphatase 
activities (A 430 units/mg protein) of particu- 
late and soluble fractions were, respec- 
tively, 37.0 ± 0.5 and 56.3 ± 7.7 for the 
isotonic NaCl group, and 44.8 ± 6.5 and 
59.5 ± 12.1 for the hypertonic NaCl group. 
The results of these experiments indicated 
that the administration of hypertonic NaCl 
did not cause a major increase or shift in the 
activities of acid phosphatase in the par- 
ticulate and soluble fractions compared to 
the corresponding fractions of the control 
group. This suggests that the lysosomes 
within the renal ceUs were probably unaf- 
fected by the administration of hypertonic 
NaCl and therefore may not be a contribu- 
tory factor for the observed increase in the 
RNase activity of the kidneys of the ex- 
perimental rats. 

Discussion. The data presented in this 
paper demonstrate that the acute adminis- 
tration of a hypertonic NaCl solution 
caused disaggregation of polyribosomes 
and inhibition of protein synthesis in the 
kidneys. In addition, the kidneys of the ex- 



perimental rats developed an increase in the 
activity of RNase. 

Earlier studies from our laboratory have 
been concerned with how the liver is af- 
fected by hypertonic solutions particularly 
relating to the polyribosomes and protein 
synthesis in the livers of mice and rats. The 
results indicated that the oral or in- 
traperitoneal administration of hypertonic 
salt solutions rapidly (within 5 min) pro- 
duced an elevated osmotic pressure of the 
portal blood (1, 3). This increase in the 
portal blood osmolarity was associated with 
disaggregation of hepatic polyribosomes 
and with a decrease in both in vitro and in 
vivo incorporation of L-[*'*C]leucine into 
hepatic and plasma proteins (1,3). Changes 
in the polyribosomal profiles and in the 
rates of protein synthesis in response to 
changes in osmotic pressure caused by ex- 
posure to hypertonicity have also been ob- 
served by others in a variety of cells and 
organisms. Robbins et al. (13) and Wengler 
and Wengler (14) have reported that HeLa 
cells placed in hyperosmotic media re- 
vealed rapid disaggregation of their poly- 
ribosomes. Hsiao (15) has reported that 
treatment with a hyperosmotic medium 
caused disaggregation of polyribosomes in 
com leaves. Chrispeels (16) also observed a 
general inhibition of protein synthesis in 
aleurone cells of barley by hyperosmotic 
media. These findings, along with our pre- 
vious data with the liver and with the pre- 
sent findings with the kidney, suggest that 
protein synthesis by different cells and or- 
gans are sensitive to alterations in the os- 
motic pressure of the extracellular fluid. 

The increased levels of RNase activity in 
the postmitochondrial supernatant fractions 



264 



HYPERTONIC NaCl AND RENAL PROTEIN SYNTHESIS 



of the kidneys of rats treated with a hyper- 
tonic NaCl solution suggests that the dis- 
aggregation of the kidney polyribosomes 
may be secondary due to the elevation in 
the levels of ribonucleases. In contrast, the 
polyribosomal disaggregation observed in 
the liver in response to the administration 
of 4% NaCl, reported in the earlier studies, 
was not found to be associated with any 
changes in the levels of RNase activity (2). 
In an attempt to consider a mechanism by 
which the administered hypertonic NaCl 
causes polyribosomal disaggregation and 
decreases in in vitro protein synthesis in the 
kidney, one must consider that the elevated 
RNase could be of importance. Elevations 
of renal RNase activity in the experimental 
rats could be due to: (1) an increase in ab- 
sorption or uptake by the kidneys of cir- 
culating RNase or (2) tissue injury causing 
lysosomal release of RNase. It has been re- 
ported earlier that rat serum normally con- 
tains a relatively high level (higher than in 
serum of mice) of RNase activity (4). Con- 
ceivably, this circulating RNase may be- 
come absorbed or extracted by the kidneys 
of rats treated with hypertonic NaCl. Our 
failure to find changes in levels of serum 
RNase activities in control and experimen- 
tal rats in the present study does not sup- 
port that the blood levels are of great im- 
portance. Yet, it is still possible that the 
kidneys of the experimental rats are able to 
extract RNase from the circulating blood 
without appreciably altering the levels of 
activities in the blood. Such a mechanism 
has been speculated as being responsible 
for the elevation of RNase in the remaining 
kidney after unilateral nephrectomy ( 17, 18) 
where, however, polyribosomes and pro- 
tein synthesis were not depressed. Also, it 
is possible that the hypertonic NaCl causes 
increased osmolarity within the circulating 
blood which induces cellular injury within 
the kidneys. Such injury could affect the 
lysosomes within renal cells and cause an 
elevation in free RNase activity. Our failure 
to find an increase or shift in acid phos- 
phatase activities of the particulate and su- 
pernatant fractions of the kidneys of control 
and hypertonic NaCl-treated rats does not 
support this possibility. 



Following the administration of hyper- 
tonic NaCl, there is an elevation in NaCl 
concentration in the liver (1,3) and proba- 
bly also in the blood and in other organs, 
including the kidneys. This raises the pos- 
sibility that an increase in the concentration 
of NaCl in the kidneys of rats treated with 
hypertonic NaCl may influence the assay of 
renal RNase activity. However, this possi- 
bility seems unlikely since the RNase ac- 
tivities in the sera of control and experi- 
mental rats were similar even though the 
NaCl concentration in the serum of the lat- 
ter was elevated over that of the former 
group. 

Further studies are needed to establish 
whether the disaggregation of kidney 
polyribosomes in hypertonic NaCl-treated 
rats is indeed due to elevated levels of 
RNase activity or to other mechanisms. 
Also, further studies are necessary to de- 
termine whether the disaggregation of 
polyribosomes and inhibition of protein 
synthesis of both liver and kidney by 
hypertonicity are mediated through similar 
or different mechanisms. 

In the past, attempts to isolate large 
polyribosomes from rat kidney have been 
unsuccessful (5, 11). The reason for the dif- 
ficulty in isolating intact polyribosomes 
from rat kidney compared to other tissue 
(e.g., rat liver or mouse kidney) is thought 
to be that rat kidney contains a high level of 
RNase activity (higher than in other tissues) 
(5, 19-21) and a low level of the naturally 
occurring RNase inhibitor (21) that stabilize 
polyribosomes (22). More recently Liu and 
Matrisian (4) reported a method to isolate 
RNase-free intact polyribosomes from rat 
kidney whereby they used an in situ arterial 
perfusion of rat kidney coupled with 
homogenization of the perfused rat kidney 
in heparin and detergents-fortified high- 
speed supernatant prepared from rat liver. 
In our hands, we were able to prepare large 
polyribosomes from kidneys of control rats 
(Fig. 1) by the employment of high concen- 
trations of Mg^^ and the omission of deter- 
gent treatment during the isolation proce- 
dure. Another possible reason for obtaining 
heavier polyribosomes with very little deg- 
radation is that we have examined the 



HYPERTONIC NaCl AND RENAL PROTEIN SYNTHESIS 



265 



polyribosomal profiles using only post- 
mitochondrial supernatants on linear su- 
crose gradients, thus eliminating several 
hours delay in the preparation of polyribo- 
somes by the procedures normally used by 
others. Furthermore, our preparations of 
kidney postmitochondrial supernatants of 
control rats exhibited much lower levels of 
RNase activity compared to the levels in 
the rat kidney homogenate reported by Liu 
and Matrisian (4). Thus, in the present 
study these modifications used in our isola- 
tion procedure may have attributed to the 
lower levels of RNase activity in the rat 
kidneys and probably resulted in the recov- 
ery of heavier polyribosomes in the kidneys 
of control rats. 



1. Lynn JK, Sarma DSR, Sidransky H. Response of 
hepatic polyribosomes of the mouse to the ad- 
ministration of anisotonic solutions. Life Sci 
10:385-394. 1971. 

2. Lynn JK, Murty CN, Sidransky H. Osmoregula- 
tion of ribosomal function in mouse liver. Biochim 
Biophys Acta 299:444-451, 1973. 

3. Lynn JK, Sidransky H. Effect of changes of os- 
motic pressure of portal blood on hepatic protein 
synthesis. Lab Invest 31:332-339, 1974. 

4. Liu DK, Matrisian PE. Isolation of ribonuclease- 
free intact polyribosomes from rat kidney. Biochim 
Biophys Acta 563:445-453, 1979. 

5. NichoUs DM, Ryan MP, Miall SH, Cappon ID. 
Factors affecting protein synthesis in rat kidney 
ribosome preparations. Canad J Biochem 
48:105-112, 1970. 

6. Rahman YE. Existence of third ribonuclease in rat 
liver particulate. Biochim Biophys Acta 119:470- 
479, 1966. 

7. Tani Y, Ogata K. Acid phosphatase having 
pyridoxine-phosphorylating activity. In: McCor- 
mick DB, Wright LD, eds. Methods in Enzymol- 
ogy. New York, Academic Press, Vol 18:p630, 
1970. 

8. Lowry OH, Rosebrough NJ, Fair AL, Randall 
RJ. Protein measurement with the Folin phenol 
reagent. J Biol Chem 193:265-275, 1951. 

9. Fleck A, Munro HN. The precision of ultraviolet 
absorption measurements in the Schmidt -Thann- 



hauser procedure for nucleic acid estimation. 
Biochim Biophys Acta 55:571-583, 1%2. 

10. Shires TK, Ekren T, Hinderaker P. Pitot HC. Ef- 
fect of ribosomal removal on rough microsomal 
composition and polysome attachment in vitro. 
Biochim Biophys Acta 374:59-75, 1974. 

11. Zomzely CE, Roberts S, Gruber CP. Brown DM. 
Cerebral protein synthesis. II. Instability of cere- 
bral messenger ribonucleic acid -ribosome com- 
plexes. J Biol Chem 243:53%-5409, 1968. 

12. Priestly GC, Malt RA. Development of the 
metanephric kidney. Protein and nucleic acid 
synthesis. J Cell Biol 37:703-715, 1%8. 

13. Robbins E, Pedcrson T, Klein P. Comparison of 
mitotic phenomena and effects induced by 
hypertonic solutions in HeLa cells. J Cell Biol 
44:400-416, 1970. 

14. Wengler G, Wengler G. Medium hypertonicity 
and polyribosome structure in HeLa cells. The 
influence of hypertonicity of the growth medium 
on polyribosomes in HeLa cells. Eur J Biochem 
27:162-173, 1972. 

15. Hsiao TC. Rapid changes in level of polyribo- 
somes in Zea-M in response to water stress. Plant 
Physiol 44:39-44, 1%9. 

16. Chrispeels MJ. Mechanism of osmotic regulation 
of hydrolase synthesis in aleurone cells of barley: 
Inhibition of protein synthesis. Biochem Biophys 
Res Commun 53:99-104, 1973. 

17. Rabinovitz M. Plasma ribonuclease uptake by rat 
kidney cortex. Proc Soc Exp Biol Med 100:865- 
867, 1959. 

18. Royce, PC. Role of renal uptake of plasma protein 
in compensatory renal hypertrophy. Amer J 
Physiol 212:924-930, 1%7. 

19. Roth JS. Ribonuclease. III. Ribonuclease activity 
in rat liver and kidney. J Biol Chem 208:181 - 194, 
1954. 

20. De Lamirande G, Allard C. Studies on the dis- 
tribution of intracellular ribonucleases. Ann NY 
Acad Sci 81:570-584. 1959. 

21. Liu DK, Matrisian PE. Regional differences in 
ribonuclease content of rat and mouse kidney. 
Biochem J 164:371-377, 1977. 

22. Blobel G, Potter VR. Relation of ribonuclease and 
ribonuclease inhibitor to the isolation of poly- 
somes from rat liver. Proc Nat Acad Sci USA 
53:1283-1288, 1966. 

Received May 17, 1982. P.S.E.B.M. 1982, Vol. 171. 



mOChtOtHO§ Of THE SOCIETY FOK EXPEUMENTAL BIOLOCY AND MEDICINE 171, 266-271 (1982) 

Classification of Hybridomas to Respiratory Syncytial Virus Glycoproteins (41509) 
BRUCE F. FERNIE/ PAUL J. COTE, Jr., and JOHN L. GERIN 

Diviiion of Molecular Virology A Immunology, Georgetown University, 5640 Fishers Lane, 

RockviUe, Maryland 20852 



Abstract. We have classified 28 hybridomas to the RS virus glycoproteins, VP66, VP84, 
by virus neutralization, RIP, and RIA tests. Without resorting to RIP, a combination of 
neutralization tests and RIA on BCH4 and RS/HO cells could be used to classify 79% of the 
hybridomas antibodies correctly. VP66 contains ms^or determinants for virus neutralization 
since 11/13 hybridoma antibodies to this protein neutralized RS virus while 0/15 hybridoma 
antibodies to VP84 neutralized the virus. 



Respiratory syncytial (RS) virus is an im- 
portant respiratory pathogen of the family 
Paramyxoviridae (1) and is a leading cause 
of lower respiratory disease in young chil- 
dren (2). Before a successful RS virus vac- 
cine can be developed, more information 
about the surface proteins is needed in 
order to fully define the host response to RS 
virus infection. Such studies have been 
hampered by the difficulty in obtaining 
purified viral proteins and monospecific 
antisera. We have investigated these prob- 
lems by developing hybridomas to glutaral- 
dehyde-fixed RS virus-infected HeLa cells. 
The initial fusion resulted in seven clones 
secreting monoclonal antibodies to three 
viral proteins (3). 

The development of Balb/c cells persis- 
tently infected with RS virus (BCH4) (4) 
provided a convenient substrate for ra- 
dioimmunoassay (RIA) and allowed us to 
eliminate clones secreting antibodies to 
human components early in the screening 
process. These cells potentially could in- 
duce antibodies better if injected into a 
syngeneic host to minimize antibody re- 
sponse to cellular antigen. We, therefore, 
ii\jcctcd Balb/c mice with syngeneic BCH4 
cells for the production of viral antigen- 
specific hybridomas. 

Although there has been a number of 
studies involved with the identification of 
RvS virus glycoproteins (5-11) the results 



I \> whom all correspondence should he addressed. 



are not straightforward and are often con- 
tradictory. Most authors present data for 
three glycoproteins with molecular weights 
of 70,000-90,000, 40,000-50,000, and 
19,000-25,000. Wunner and Pringle (11), 
however, could not confirm the presence of 
the 70,000-90,000 or 19,000-25,000 mol wt 
glycoproteins, but instead found two glyco- 
proteins, mol wt 40,000-50,000 (VGP48 
and VGP42). Recently the large glycopro- 
tein was immunoprecipitated from infected 
cells (5, 6) but other glycoproteins, mol wt 
45,000-50,000 and 17,000-24,000, were 
not. The development of monoclonal an- 
tibodies with specificity for the RS virus 
glycoproteins would greatly help in the 
identification and study of these glycopro- 
teins. 

Materials and Methods. Cells and vi- 
ruses. HeLa Ohio (HO) cells were obtained 
from Flow Laboratories (McLean, Va.). 
The Balb/c cell line (BCH4), persistently 
infected with RS virus, has been described 
(4). Both cell lines were grown in equal 
parts of minimum essential medium with 
Earles salts and basal Eagle medium with 
Hank s salts, 10% heated (56745 min) fetal 
bovine serum (Dutchland Labs) and 4 mAf 
glutamine. Suspension cultures of HO cells 
were grown in 10% heated fetal bovine 
serum, 4 mM glutamine in Eagles minimum 
essential medium, spinner modified. HO 
cells were infected (RS/HO) in suspension 
with 1 PFU/cell of RS virus. Long strain. 
The cells were rinsed 2 hr after infection 
and resuspended in suspension medium, 
with 5% fetal bovine serum, at 2 x 10* 



266 
/«ffj h% ifw SiK'iet\ /cv h:tpcnmcnt»l Biology and Medicine. 



HYBRIDOMAS TO RS VIRUS GLYCOPROTEINS 



267 



cells/ml. RS/HO cells were labeled with 10 
/xCi/ml of D-[l,6-^H]glucosamine hydro- 
chloride (New England Nuclear, sterile 
aqueous solution) 12 hr postinfection. Cells 
and culture fluid were collected 24 hr post- 
infection. Virus-free culture fluid was pre- 
pared by peUeting the virus at 7500 rpm for 
4 hr in a Beckman J-21C centrifuge (12). 

Virus neutralization. Hybridoma culture 
fluids were tested for virus neutralizing 
antibodies as described (3) with minor 
modifications. Four units of guinea pig 
complement (Flow Laboratories) were 
added throughout to increase the sensitivity 
of the assay (13). 

Indirect immunofluorescence (JFL), IFL 
was performed using live RS/HO cells as 
described (4). 

Radioimmunoprecipitation (RIP). RIP 
using iodinated viral proteins (14), pH]- 
glucosamine-labeled cell lysate, or virus- 
free culture fluid from pH]glucosamine-la- 
beled RS/HO cells was performed as de- 
scribed (14). Samples were analyzed by slab 
gel electrophoresis (14). 

Radioimmunoassay (RIA). Indirect RIA 
using ^^^I-goat anti-mouse y globulin as the 
second antibody and methanol-fixed RS/ 
HO or BCH4 cells as the substrate was 
recently described (3). 

Somatic cell hybridizations. The genera- 
tion of hybridomas followed standard tech- 
niques as described for RS virus, except vi- 
able BCH4 cells were used as the immuno- 
gen instead of glutaraldehyde-fixed RS/HO 
ceUs (3). 

Results. Approximately 1000 hybridoma 
colonies were originally screened by RIA 
using methanol-fixed BCH4 cells. Ten per- 
cent appeared to be RS virus specific; these 
colonies were cloned and virus neutraliza- 
tion, RIP, IFL, and further RIA analysis 
using methanol-fixed BCH4 and RS/HO 
cells performed on the spent culture fluid. 

Immunofluorescence. All of the 28 clones 
chosen for further study had the staining 
characteristics of surface proteins as de- 
termined on live RS/HO cells (4). Typical 
results are shown in Fig. 1. 

Radioimmunoprecipitation. RIP analysis 
indicated that about 75% of the clones pre- 
cipitated one of two viral glycoproteins 
(Fig. 2, Table I); designated VP84 and VP66 




Fig. 1. Living cell immunofluorescence of RS/HO 
cells. RS/HO cells were prepared for living cell immu- 
nofluorescence as described (4) 22 hr after infection. 
The cells were incubated with (A) monoclonal 111, 
class 84 A; (B) monoclonal 13-1 class 66 A. 



(14). VP84, a protein not present in great 
abundance on the virion (7, 14) has been 
identified using RIP (14). VP66 consists of 
two disulfide-bonded polypeptides with 
molecular weights of 43,000 and 19,000 
(14); both polypeptides are glycosylated, 
VP19 relatively more so than VP43. 
Seventy-seven percent of the clones spe- 
cific for VP66 could precipitate iodinated 
viral VP66 protein while only 1 of 15 (6%) 
clones specific for VP84 could precipitate 
iodinated VP84 (Table I). Seventy-three 
percent of the clones specific for VP84, 
however, precipitated pH]glucosamine- 
labeled VP84 (Table I). RIP using pH]- 
glucosamine-labeled RS/HO cell lysates 
showed the presence of VP66 (VP43 and 
VP19) when precipitated by monoclonal 
antibodies specific for VP66. However, 
VP84 and an additional glycoprotein with 
a mol wt of 43,000 to 46,000 precipitated 
from cell lysates by monoclonal antibodies 
specific for VP84. This protein, which is 
not readily detected in virus-free culture 
fluid, may represent a cleavage product 
of VP84 (Fig. 2). The width and very 
diffuse nature of the protein band is very 
similar to that obtained with VP84, but 
markedly different from VP43, indicating 
that VP43 has not coprecipitated with 



268 



HYBRIDOMAS TO RS VIRUS GLYCOPROTEINS 



4I-OLU0O8AMINE LABELED 
R8/HOCELLLY8ATE 




VIRUS FREE CULTURE FLUID 



CO 

O 

< 
o 
z 
< 



i S 



< O ID 

CO CO CD 



200K 
94 

68 



43 




25.7 



18.4 



14.3 

Fui. 2. RIP of RS virus-specific glycoproteins from cell lysates or culture fluid. (A) RS/HO cells 
grown in suspension were labeled from 12 to 24 hr postinfection with ['HJglucosamine. The cells were 
pelleted* rinsed twice in Hank's balanced salts, 50 mM Hepes. pH 7.5. PMSF 100 mg/ml, and sus- 
pended in the same buffer at 5 x 10^ cells/ml. Cell lysates were prepared by adding Triton X-100 to 
0.1% and incubating at 3T for 30 min. Nuclei and large membrane fragments were removed and the 
supernatant made up to 0.1% SDS. This mixture was incubated at 45° for 30 min. Approximately 5 x 
10^ cell equivalents of this cell lysate were incubated with 100 fi\ of washed Immunoprecipitin 
(Bethesda Research Labs.), cleared, and incubated with the indicated antibody. The exact RIP proce- 
dure has been described (14). Samples for electrophoresis were run on slab gels consisting of 14% 
ucrylamide with DATD as the crosslinker (14), using the discontinuous system of Laemmli (15). The 
specific antibodies used were as follows: class 66A, RS 13-1; class 84B, RS 18-1: class 84A, RS 111; R 
unli-I-CS. rabbit anti-fetal calf serum (14); '"I RS, iodinuted RS virus proteins precipitated with rabbit 
nnti-RS virus (14). '^M-VP84 is present in immunoprecipitatcs with rabbit anti-RS virus (14) but is not 
visible in this photograph. 

(B) A vims-free culture fluid preparation containing only VP84 was used to screen for monoclonals 
to this protein. RIP and gel electrophoresis were as described (14). The specific antibodies used were 
as follows: class 66A. RS 13-1: class 84A, RS 111: class 84C. RS 322: class 84B, RS 821. 



VP84. Although Staphylacoccal protease 
treatment of culture fluid containing VP84 
generates a fragment with a molecular 
weight of 49,000 (S. Spring, personal com- 



munication), we have not eliminated the 
possibility that the coprecipitating protein 
is due to cellular contamination. If the 
40,000-45,000 mol wt protein precipitated 



HYBRIDOMAS TO RS VIRUS GLYCOPROTEINS 



269 



TABLE I. Classification of Hybridomas to 
RS Virus Surface Proteins 




RIP* 


Relative 
binding 
ratio*^ 


No. of 
clones 


Class 


Neut.« V S 



66A 
66B 
6oC 
66D 
66E 

84A 
84B 

84D 



+ 
+ 
+ 



+ 
+ 

+ 

+ + 

+ 



<1 
1-2 

<1 
<1 

2 
<2.5 
3-5 

2 



' Viral neutralization activity scored positive if at 
least 50% inhibition is found at 1:20 dilution. 

* RIP performed using: iodinated viral proteins (V), 
['Hlglucosamine-labeled, virus-free culture fluid from 
RS/HO cells (S). This fluid was shown to be free of 
pHlglucosamine-labeled VP66 by RIP/SDS-PAGE. 
Viral protein precipitated is indicated by class number. 
See Fig. 2. 

*" Relative binding ratio is determined from com- 
parative RIA on RS/HO and BCH4 cells per Fig. 3. 



by anti-VP84 antibodies is a cleavage prod- 
uct, then it probably represents the VGP48 
of Wunner and Pringle (11) while VP43 is 
their VPG42. This interpretation is based 
on the relative level of pH]glucosamine in- 
corporation into VP43 and VP19 compared 
to the relative level of pH]glucosamine into 
proteins with mol wt of 40,000-50,000 and 
19,000-24,000, respectively (5-8, 11). 

Virus neutralization. All of the selected 
clones have been tested for virus neu- 
tralization activity (Table I). None of the 
monoclonal antibodies with neutralizing 
activity appears to be directed against 
VP84. Lack of neutralizing activity was not 
caused by low antibody concentration as 
indicated by the level of antibody binding in 
the RIA tests as shown below. 

Radioimmunoassay. When clones were 
compared by RIA using BCH4 or RS/HO 
cells as substrate consistent patterns in the 
P/N binding ratios were observed (Fig. 3). 
These differences were used to classify 
clones as VP84-like or VP66-like (3). This 
has been useful but not entirely accurate 
since there is some overlap between 
monoclonal groups specific for the two 




12 3 4 



Fig. 3. Characterization of monoclonal antibodies 
to RS virus glycoproteins by RIA. Assay of antibodies 

with anti-VP66 activity ( ) on RS/HO (A) and 

BCH4 (B) cells; monoclonal antibody and class iden- 
tifications in (A) and (B) are (•) RS 284:66A, (O) RS 
334:66B, (A) RS 162:66C, (A) RS 47:66D, and (■) RS 
184:66E. Assay of antibodies with anti-VP84 activity 

( ) on RS/HO (C) and BCH4 (D) cells; monoclonal 

antibody and class identifications in (C) and (D) are 
(•) RS 1 1 1:84A, (O) RS 93:84B, (A) RS 104:84C, and 
(A) RS 402:84D. P/N is the cpm for the culture media 
samples to that for blank culture media (100-175 
cpm). Antibodies assayed on uninfected HeLa or 
Balb/c cells exhibited P/N values <2. (E) The relative 
binding ratio was determined by dividing the P/N 
number obtained on RS/HO cells by the P/N number 
obtained on BCH4 cells. 



surface proteins. In addition, monoclonals 
specific for RNP exhibit binding ratios sim- 
ilar to those for VP66 (data not shown). 
Further testing will be required to eliminate 
these problems. 



270 



HYBRIDOMAS TO RS VIRUS GLYCOPROTEINS 



DiscussioiL A fusion using spleenocytes 
from mice immunized with live BCH4 ceUs 
yielded approximately 15 stable clones se- 
creting monoclonal antibodies to each of 
two surface proteins. We developed a clas- 
sification scheme for these hybridomas 
based on IFL, virus neutralization, RIP and 
RIA. This classification scheme divides 
monoclonals to VP84 into four subgroups 
and those to VP66 into five subgroups. In 
most cases the assignment to a particular 
subgroup was straightforward. However, a 
small portion of the clones that failed to 
specifically precipitate a viral protein have 
been assigned to a subgroup based on RIA 
data alone and these assignments must be 
regarded as tentative. As knowledge about 
the importance of each subgroup grows, it 
should be possible to rapidly screen new 
fusions for the desired hybridomas. 

In the interest of developing a rapid, sim- 
ple screening technique for hybridomas to 
RS virus surface proteins, a variety of tests 
have been performed on the 28 hybridomas 
isolated to date. If initial screening is done 
using BCH4 and RS/HO cells and this is 
followed by virus neutralization tests, 79% 
of the selected clones can be placed in their 
correct classification. This procedure 
eliminates the need to screen each initial 
clone by the cumbersome procedures of 
IFL and RIP. 

Wc are currently studying reasons for the 
differential binding of some classes of 
monoclonal antibodies to BCH4 and 
RS/HO cells. This may simply reflect dif- 
ferences in the amount of VP84/VP66 in 
these cells or differences in the orientation 
of these proteins on the cell surface. 

The efficacy of using BCH4 cells as im- 
munogen and substrate for the production 
and screening of hybridomas to RS virus 
has been established. The success obtained 
using BCH4 cells as an immunogen proba- 
bly stems from the lack of response to 
BCH4 cellular antigens. In addition, the 
persistently infected BCH4 cells are pro- 
viding a continual source of viral antigens 
for purification and characterization. 

VP66 contains major determinant(s) in- 
volved in virus infectivity, VP84 apparently 
does not. We have not been able to assign a 



definite function to VP84, although this 
protein has the physicochemical character- 
istics of the paramyxovirus hemaggluti- 
nin (14). 

The use of monoclonal antibodies with 
specificity for VP84 and VP66 should help 
clarify the complex picture of RS virus gly- 
coproteins. The fact that two antigenicaUy 
distinct proteins have very similar molecu- 
lar weigiht raises some question as to the 
identity of the 45,000-50,000 mol wt glyco- 
protein found in nonimmunoprecipitated 
RS virus or infected ceUs (7-11). In fact, 
the protein that coprecipitates with VP84 
may represent VGP48 as defined by Wun- 
ner and Pringle (11) while VP43 may repre- 
sent their VGP42. Earlier results obtained 
by direct analysis of viral proteins has not 
yet been consolidated with more recent 
work involving the RIP of viral proteins (5, 
6, 14). This is caused, in part, by the vari- 
able precipitation efficiency of RS virus 
proteins (5, 14) resulting in a selection of 
viral proteins that may or may not be pres- 
ent in great abundance. The selection pro- 
cess, coupled with the presence of viral 
proteins having similar molecular weight 
(e.g., VGP48, VGP42; 11), VP43, postu- 
lated cleavage product of VP84 (this paper) 
has produced considerable uncertainty 
about the number and size of the RS virus 
glycoproteins (5-11). This uncertainty has 
prevented a comprehensive understanding 
of RS virus glycoproteins from being devel- 
oped. Hopefully, this situation will be 
clarified as monoclonal antibodies are used 
to examine the glycoproteins of RS virus 
more extensively. 



The authors wish to thank E. Ford and G. Dapolito 
for their technical assistance and V. Reap for her sec- 
retarial assistance. 

This work was supported in part by Contract NOl- 
A 1-22665 between Georgetown University and the 
NIAID. 



1. Kingsbury DW, Bratt MA, Choppin PW, Hanson 
RP, Hosaka Y, ter Meulen V, Norrby E, Plow- 
right W, Rott R, Wunner WH. Paramyxoviridae. 
Intervirology 10:137-152, 1978. 

2. Jacobs JW, Peacock DB, Comer BD, Caul EO, 
Clarke KR. Respiratory syncytial and other vi- 



HYBRIDOMAS TO RS VIRUS GLYCOPROTEINS 



271 



nises associated with respiratory disease in in- 
fants. Lancet 1:871-876, 1971. 

3. Cote, Jr PJ. Femic BF. Ford EC, Shih JW-K, 
Gerin JL. Monoclonal antibodies to respiratory 
syncytial virus: Detection of virus neutralization 
and other antigen-antibody systems using infected 
human and murine cells. J Virol Methods 
3:137-147. 1981. 

4. Femle BF. Ford EC. Gerin JL. The development 
of Balb/c cells persistently infected with respira- 
tory syncytial virus: Presence of ribonucleopro- 
tein on the cell surface. Proc Soc Exp Biol Med 
167:83-86, 1981. 

5. Bernstein JM, Hruska JF. Respiratory syncytial 
virus proteins: Identification by immunoprecipi- 
tation. J Virol 38:278-285, 1981. 

6. Dubovi EJ. Analysis of proteins synthesized in 
respiratory syncytial virus-infected cells. J Virol 
42:372-378. 1982. 

7. Levine S. Polypeptides of respiratory syncytial 
virus. J Virol 21:427-431, 1977. 

8. Peeples M, Levine S. Respiratory syncytial virus 
polypeptides: Their location in the virion. Virol- 
ogy 95:137-145, 1979. 

9. Pringle CR. Shirodaria PV, Gimenez HB, Levine 
S. Antigen and polypeptide synthesis by 



temperature-sensitive mutants of respiratory syn- 
cytial virus. J Gen Virol 54:173-183, 1981. 

10. Ueba O. Respiratory syncytial virus. II. Isolation 
and morphology of the glycoproteins. Acta Med 
Okayama 34:245-254. 1980. 

11. Wunner WH, Pringle CR. Respiratory syncytial 
virus proteins. Virology 73:228-243, 1976. 

12. Femie BF, Gerin JL. The stabilization and purifi- 
cation of respiratory syncytial virus using MgS04. 
Virology 106:141-144, 1980. 

13. Buynak EB, Weibel RE, Carlson AJ, McLean 
AA, Hilleman MR. Further investigations of live 
respiratory syncytial virus vaccine administered 
parenterally. Proc Soc Exp Biol Med 160:272- 
277, 1979. 

14. Femie BF, Gerin JL. Immunochemical identifica- 
tion of viral and non-viral proteins of the respira- 
tory syncytial virus virion. Infect Immun 
37:243-249, 1982. 

15. Laemmli UK. Cleavage of structural proteins 
during the assembly of the head of Bacteriophage 
T4. Nature (London) 227:680-685, 1970. 



Received May 17, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 272-275 (1982) 



Oligoclonal IgG in the Cerebrospinal Fluid of Guinea Pigs with 
Experimental Allergic Encephalomyelitis (41510) 

M. IIVANAINEN,*^ B. DRISCOLL,t J. RICHERT,* M. LEON,* A. CHU,* 
M. KIES,t B. BROWN,* W. WALLEN,* D. MADDEN,* and J. SEVER* 

* Infectious Diseases Branch, National Institute of Neurological and Communicative Disorders and Stroke: 
^Laboratory of Cerebral Metabolism, National Institute of Mental Health: and XNeuroimmunology Branch, 
National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health, 

Bethesda, Maryland 20205 



Abstract. Oligoclonal IgG bands were demonstrated in the cerebrospinal fluid of guinea 
pigs with experimental allergic encephalitis before clinical signs developed. No bands were 
found in the serum of these animals nor were bands found in the CSF or serum of control 
animals. Immunofixation techniques demonstrated that the bands were immunoglobulins of 
the y G class. 



Experimental allergic encephalitis (EAE) 
is an acute neurological disease which can 
be produced in guinea pigs, rats, and other 
animals by injection of brain tissue or ex- 
tracts from brain tissue such as myelin 
basic protein in complete Freund's adju- 
vant. The disease has been proposed as a 
possible experimental model for multiple 
sclerosis (MS) (1). The presence of oligo- 
clonal IgG bands in the cerebrospinal fluid 
(CSF) of patients with MS is one of the 
most common immunological findings in 
this disease (2). Most patients with optic 
neuritis, acute herpes encephalitis, Ep- 
stein -Barr encephalitis, subacute scle- 
rosing panencephalitis, and neurosyphilis 
also have bands present (3-5). Oligoclonal 
IgG is present in some patients with Guil- 
lain-Barre, myasthenia gravis, and Par- 
kinson's disease but is not present in normal 
individuals (3, 6, 7). We report here the 
occurrence of oligoclonal IgG in the spinal 
fluid of guinea pigs with EAE. 

Materials and Methods. Guinea pig my- 
elin basic protein (BP) was prepared as pre- 
viously described (8). Adult strain 13 guinea 
pigs were sensitized by injection of 0.7 mg 
BP in complete Freund's adjuvant (3.5 mg 
of heat-killed mycobacteria (HayRv) total) or 
0.1 mg BP in complete Freund's adjuvant 



' To whom all correspondence should be addressed. 



(0. 1 mg of heat killed mycobacteria) in five 
sites in the nuchal region and in each hind 
foot pad. Injection of this material causes 
EAE in 100% of strain 13 guinea pigs within 
21 days. At the time CSF was obtained 
(Day 11) none of the guinea pigs had devel- 
oped clinical signs of disease and no cir- 
culating antibodies to BP were detectable 
(9). Control animals were injected with 
either the same emulsion containing no BP 
or with an emulsion containing BP but no 
mycobacteria. None of the control animals 
displayed any clinical signs of EAE. 

The guinea pigs were anesthetized using 
ether inhalation for cisternal puncture. The 
cisternal subarachnoid space was entered 
with a 23-gauge needle and the sample was 
allowed to fill the needle. CSF samples 
from control and EAE animals were ccn- 
trifuged in a Beckman microfuge at 10,000 
rpm for 2 min and carefully transferred to 
conical plastic vials and stored at -20**; 
only samples with no evidence of hemolysis 
or red cells were included in this study. 
Blood was obtained by cardiac puncture 
and was allowed to clot at room tempera- 
ture. The sample was centrifuged and 
serum was stored at -20°. 

A modification of the SDS- poly aery 1- 
amide gel electrophoresis was used for 
the determination of oligoclonal IgG (10). 
With this new method it is possible to test 
unconcentrated CSF. Briefly, the electro- 



272 
OOJ7-9727/82n0272-04$0l.00/0 

igftt ,cy 1982 hy the Society for Experimental Biology and Medicine. 
*ts renerved. 



CSF OLIGOCLONAL IgG BANDS IN EAE 



273 



phoresis was carried out in vertical gel 
slabs. Acrylamide content of the stacking 
gel was 5% and the separating gel was 9%. 
Samples (50 fil of unconcentrated CSF or 
serum diluted 1:200) were prepared for 
electrophoresis by mixing 10 parts CSF or 
diluted serum with one part buffer (70% su- 
crose, 10% SDS, and bromophenol blue as 
dye). These specimens were placed directly 
on the stacking gel with no further treat- 
ment. We tested CSF samples from EAE 
and control animals with comparable IgG 
concentrations, i.e., 3 fig of IgG contained 
in 50 fil CSF. Oligoclonal bands were de- 
tected in EAE animals while no bands were 
found in control animals. These results in- 
dicate that the appearance of oligoclonal 
bands is independent of the IgG concentra- 
tion. Purified guinea pig IgG were used as 
controls. After electrophoresis with a con- 
stant current of 10 to 30 mA per gel for 3-4 
hr, the slabs were removed and stained for 
protein with Coomassie brilliant blue. Two 
or more bands in the IgG region were con- 
sidered positive. 

Immunofixation of IgG was done on the 
gel surface. Strips of cellulose acetate 
membrane were impregnated by soaking in 
the appropriate undiluted antiserum (anti- 
guinea pig IgG heavy and light chains, H & 
L) prepared in goat (Cappel Laboratories). 
The gels were overlaid with the strips for 2 
hr in a moisture chamber at 4**. Antibody 
excess was carefully washed out overnight 
in phosphate-buffered saline (PBS) (pH 7.2) 
at room temperature. To detect the reaction 
between the oligoclonal IgG and specific 
immune sera, the washed gel was incubated 
with peroxidase conjugated rabbit anti-goat 
IgG (Miles- Yeda LTD, Israel) for 2 hr in a 
moisture chamber at 4*". The gel was 
washed overnight with frequent changes of 
PBS (pH 7.4) and reacted with diaminoben- 
zidine (11). The enzymatic reaction was 
stopped by washing with cold tap water 
when color developed in the gel. The dark 
brown bands demonstrated by this tech- 
nique confirmed the presence of IgG bands. 
Side-by-side comparisons were done on 
identical samples in which one lane was 
stained with Coomassie brilliant blue and 
another lane with peroxidase-conjugated 
antisera, to prove that IgG was present. 



Results. A modified SDS- poly acryl- 
amide electrophoresis method was used 
to detect oligoclonal bands in the uncon- 
centrated CSF of guinea pigs in which EAE 
was induced. The animals were clinically 
well at the time they were studied. With the 
sensitizations used clinical signs of EAE 
are observed by 21 days after inoculation. 
Oligoclonal IgG bands were observed in the 
CSF of all 14 guinea pigs obtained 1 1 days 
after inoculation with 0.7 mg BP (Table I). 
Three bands were detected in 5 and two 
bands in 9 guinea pigs using this concentra- 
tion (Fig. 1). Two oligoclonal bands were 
also found in three of four guinea pigs 
which were inoculated with 0.1 mg BP. 
These results suggest that the variation in 
BP concentrations did not affect the ap- 
pearance of oligoclonal bands in these ani- 
mals. None of the control guinea pigs had 
detectable components in the IgG region of 
the CSF. 

The electrophoretic pattern of the sera of 
the inoculated guinea pigs showed no dif- 
ference when compared to the control 
groups; all of the sera were considered to 
have no oligoclonal IgG (Table I). 

The globulin class of the oligoclonal 
bands was determined by immunofixation 
with purified anti-guinea pig IgG heavy and 
light chain-specific sera. The localization of 
the color in the approximate IgG area was 
observed in all samples showing specific 
oligoclonal IgG patterns (Fig. 1); those 
without oligoclonal IgG were negative. This 
confirmed the class of the oligoclonal IgG. 

Discussion. Oligoclonal IgG bands were 
demonstrated in the CSF but not in the sera 
from 17 of 18 guinea pigs inoculated with 
myelin basic protein in complete Freund's 
adjuvant. The observation that oligoclonal 
IgG did not occur in the CSF of any of the 
control animals shows an association with 
EAE. Staining with anti-guinea pig IgG 
heavy and light chains confirmed the spec- 
ificity of the bands as IgG. The appearance 
of oligoclonal IgG in the CSF of basic 
protein-sensitized guinea pigs before the 
onset of EAE suggests that this antibody 
may contribute to the pathogenesis of the 
disease. Similar oligoclonal bands were 
found in three of four guinea pigs and re- 
ported in an abstract by Rostami (12). 



274 



CSF OLIGOCLONAL IgG BANDS IN EAE 



TABLE I. Development of Oligoclonal Bands in CSF of Guinea Pigs Sensitized with 
Basic Protein in Complete Freund's Adjuvant 





GP sensitization 
BP (mgVMTbc (mg) 


No. of 
animals 


GPwith 
oligoclonal bands 




CSF Sera 


EAE 
Controls 


0.7/3.5 
0.1/0.1 

0^0 

0^3.5 
0.7/0 
0.1/0 


CFA 
CFA 

CFA 
IFA 
IFA 


14 
4 

1 

6 
7 
2 


14 
3 ND 








Note. Abbreviations used: CFA, complete Freund*s adjuvant; IFA, incomplete Freund*s adjuvant; BP, basic 
protein; GP, guinea pig; MTbc, mycobacteria; ND, not done. 



Oligoclonal bands were also detected in 
brain extracts and CSF of guinea pigs with 
chronic relapsing experimental allergic en- 
cephalitis (R-EAE) (13). It will be important 
to determine the antigenic specificity of the 



a 



t$ 




Fig. 1. Oligoclonal IgG in CSF of guinea pigs in- 
oculated with basic protein in complete Freund's adju- 
vant (BP-CFA). SDS-PAGE electrophoresis of CSF, 

(a) CSF of guinea pig inoculated with 0.7 mg BP-CFA; 

(b) CSF of control guinea pig; (c) CSF of guinea pig 
inoculated with 0.1 mg BP-CPA; (d) immunofixation 
with an indirect peroxidase assay in CSF from guinea 
pig inoculated with BP-CFA. Note: The discrete 
oligoclonal bands (arrows) are in the IgG area. The IgG 
concentrations of CSF samples a, b, c, were similar, 
i.e., 3 /xg IgG contained in 50 /xl CSF. 



oligoclonal IgG in the sensitized guinea pigs 
since this finding may clarify the possible 
role of the IgG in these animals and may 
contribute to an understanding of the im- 
portance of oligoclonal IgG in MS. 

A recent study of CSF and serum from 
rabbits with EAE showed oligoclonal IgG 
bands in some of these animals (14). The 
animals were tested prior to sensitization 
and at the time of development of clinical 
disease. A few rabbits had bands in their 
CSF and/or serum prior to sensitization. Of 
1 1 rabbits given whole nervous tissue, 7 de- 
veloped new bands in their serum and CSF. 
The absence of bands in the CSF of some of 
the animals in these studies and presence of 
bands in the serum may relate to the dif- 
ferent species of animals studied or the 
collection of specimens at the time of ap- 
pearance of symptoms. Our animals were 
sampled prior to development of symptoms 
and bands were always present in the CSF 
but not the serum. Serial studies of sen- 
sitized animals will have to be conducted to 
determine the time of appearance of bands. 
Another study (15) demonstrated anti-BP 
activity in the oligoclonal bands in serum 
and CSF from rabbits with EAE. 

The use of the modified stacking poly- 
acrylamide gel electrophoresis permitted 
the tests in this study to be carried out with 
50-/Ltl samples of unconcentrated CSF ob- 
tained by cisternal puncture. In the prepa- 
ration of specimens for electrophoresis, the 
procedure avoided disruption of the IgG 
molecules to subunits as well as prevented 
IgG from aggregating during storage. This 



CSF OLIGOCLONAL IgG BANDS IN EAE 



275 



was done by mild treatment with SDS 
which did not break it down to the con- 
stituent light and heavy chains. Bands are 
only present in the IgG region. This was 
further confirmed by using marker proteins 
of known molecular weight and by im- 
munofixation with antibodies to guinea pig 
IgG heavy and light chains. The concentra- 
tions and defined porosity of the stacking 
and separating gels used for these studies 
did not permit entry of dimeric IgG (~ mol 
wt = 340,000 daltons) or IgM (~ mol wt = 
900,000 daltons). The oligoclonal bands 
were the result of increased production of 
IgG by certain clones of plasma cells. The 
bands appear as distinct lines, sometimes 
with a background of diffuse polyclonal 
IgG. The SDS treatment modifies the IgG 
to permit it to enter the gel and migrate as a 
single entity. The IgG, however, is not irre- 
versibly denatured because it can be eluted 
from gels without loss of functional capa- 
bility. Thus, this procedure should be use- 
ful to study small serial samples from indi- 
vidual animals. 



1. Patcrson PY, Day ED, Whitacre CC. Neuro im- 
munologic diseases: Effector cell responses and 
immunoregulation mechanisms. Immunol Rev 
55:90, 1981. 

2. Johnson KP, Arrigo SC, Nelson BJ, et al. Agarose 
electrophoresis of cerebrospinal fluid in multiple 
sclerosis: A simplified method for demonstrating 
cerebrospinal fluid oligoclonal immunoglobulin 
bands. Neurology 29:273-277. 1977. 

3. Link H, Muller R. Immunoglobulins in multiple 
sclerosis and infections of the nervous system. 
Arch Neurol 25:326-344, 1971. 

4. Link H, Norrby E, Olsson J. Immunoglobulins 
and measles antibodies in optic neuritis. N Engl J 
Med 289:1103-1104, 1973. 

5. Olsson JE, Link H. Immunoglobulin abnor- 
malities in multiple sclerosis. Relation to clinical 



parameters: Exacerbations and remissions. Arch 
Neurol 28:392-399, 1973. 

6. Laterre EC, Callewaert A, Heremans JF, et al. 
Electrophoretic morphology of gamma globulins 
in cerebrospinal fluid of multiple sclerosis and 
other diseases of the nervous system. Neurology 
20:982-990, 1970. 

7. Williams A, HouflT S, Lees A, Calne D. Oligo- 
clonal banding in the cerebrospinal fluid of pa- 
tients with postencephalitic Parkinson's. J Neurol 
Neurosurg Psychiatry 42:790, 1979. 

8. Diebler GE, Martenson RE, Kies MW. Large 
scale preparation of myelin basic protein from 
central nervous tissue of several mammalian 
species. Prep Biochcm 2:139-165, 1972. 

9. Kies MW, DriscoU BF, Lisak RP, Alvord EC. 
Immunologic activity of myelin basic protein in 
strain 2 and strain 13 guinea pigs. J Immunol 
115:75-79, 1975. 

10. livanainen MV, Wallen W, Leon ME, et al. Mi- 
cromethod for detection of oligoclonal IgG in un- 
concentrated CSF by polyacrylamide gel electro- 
phoresis. Arch Neurol 38:427-430, 1981. 

11. Graham RC, Kamovsky MF. The early stage of 
absorption of injected horseradish peroxidase in 
the proximal tube of the mouse kidney: Ul- 
trastructural cytochemistry by a new technique. J 
Histochem Cytochem 14:291-302, 1966. 

12. Rostami A, Lisak RP, Blanchard N, Guerrero F, 
Zweiman B, Pleasure D. Oligoclonal IgG in cere- 
brospinal fluid (CSF) of animals with experimen- 
tal allergic encephalomyelitis (EAE). Neurology 
31 (2): 126- 127, 1981. 

13. Mehta P, Pansmann H, Wisniewski H. Im- 
munologic studies of chronic relapsing EAE in 
guinea pigs: Similarities to multiple sclerosis. J 
Immunol 127:334, 1981. 

14. Whitacre CC, Mattson DH, Paterson PY, Roos 
RP, Peterson DJ, Amason BGW. Cerebrospinal 
fluid and serum oligoclonal IgG bands in rabbits 
with experimental allergic encephalomyelitis. 
Neurochem Res 6:87-%, 1981b. 

15. Petrasek J, Panitch H, Norrby E. Formation of 
oligoclonal antibody in experimental allergic en- 
cephalomyelitis. Neurology 31 (2): 126, 1981. 

Received April 23, 1982. P.S.E.B.M. 1982, Vol. 171. 



PROCEEDINGS OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY AND MEDICINE 171, 276-284 (1982) 



The Effect of Dehydroepiandrosterone on Adipose Tissue Cellularity in 

IVIice (41511) 

MARGOT P. CLEARY,*'^ ROBIN SEIDENSTAT,* ROBERT H. TANNEN,t 
AND ARTHUR G. SCHWARTZt 

^Department of Nutrition and Food Sciences, Drexel University, Philadelphia, Pennsylvania 19104, and 
tFels Research Institute, and Department of Microbiology, Temple University Medical Center, 

Philadelphia, Pennsylvania 19104 



Abstract. The effect of dehydroepiandrosterone (DHEA) on adipose tissue growth in 
young adult male mice was investigated. In Experiment 1, mice were individually caged and 
treated by intraperitoneal injections of DHEA (25 mg/kg body weight, 3x weekly) for 20 
weeks. No significant differences in body weight and adipose tissue growth were found. In 
Experiment 2, DHEA was administered by intubation (450 mg/kg body weight, 3x weekly). 
In addition, mice were housed either one/cage or five/cage. After 10 weeks of treatment 
body weight and epididymal and retroperitoneal fat pad weights were significantly decreased 
in both DHEA groups compared to their respective control groups. In both adipose depots 
fat cell size was significantly decreased in treated mice while there was no effect on fat cell 
number. No differences were found in adipose tissue lipoprotein lipase or liver fatty acid 
synthetase of treated mice. 



The 17-ketosteroid, dehydroepiandros- 
terone (DHEA), has been shown by Yen et 
aL (1) and Schwartz (2) to decrease body 
weight gain in yellow obese mice. This al- 
teration in body weight occurred without a 
concomitant change in food intake. It was 
also shown that DHEA treatment de- 
creased total body lipids and the rate of 
liver lipogenesis (1). In addition, Yen et aL 
(1) reported that body weight was de- 
creased in lean mice but the effect was not 
as dramatic as that seen in the obese mice. 
No additional information on lean mice was 
reported. 

DHEA is known to inhibit the enzyme 
glucose-6-phosphate dehydrogenase (3-6). 
Inhibition of glucose-6-phosphate dehydro- 
genase would theoretically result in a de- 
creased availability of NADPH for either 
lipogenesis or DNA synthesis. The studies 
in obese mice suggested that DHEA is 
primarily affecting lipogenesis. However, 
DHEA has also been shown to inhibit DNA 
synthesis in cultured cell lines (7) and in 
mouse epidermis in vivo (8) and to decrease 
the development of spontaneous breast 



' To whom all correspondence should be addressed. 



cancer formation in C3H mice (2), indicat- 
ing it can also affect cellular proliferation. It 
has been established that adipose tissue, 
the major site of stored body lipids, can 
change by either the size of the fat cells or 
the number of fat cells (9, 10). Therefore, 
the specific effects of chronic DHEA ad- 
ministration on adipose tissue growth were 
investigated in mice. Two modes of admin- 
istration for DHEA were also studied. 

Materials and Methods. Animals. Six- 
week-old male C3H mice were purchased 
from Jackson Laboratories, Bar Harbor, 
Maine. Mice were housed individually in 
wire mesh cages, unless otherwise indi- 
cated, and were allowed food and water ae/ 
libitum, A 12-h light -dark cycle was 
maintained and the temperature was kept at 
74°F. Body weight and food intake were re- 
corded weekly. 

Experiment I. Mice were housed indi- 
vidually as described and were fed Purina 
powdered rodent chow (5001, 5% fat by 
weight). One-half of the mice (n = 6/group) 
received intraperitoneal injections 3x 
weekly with 25 mg/kg body weight DHEA 
(Sigma Chemical Co., St. Louis, Mo.) in a 
homogenized suspension of saline: Emul- 
phor (GAP, New York, N.Y.) (95:5, v/v). 



276 
OOJ7-9727/82/1 10276-09$0l .00/0 
Copyright ,r) 1982 by the Society for Expcrimentai Biology and Medicine. 
^VntJt reserved. 



DEHYDROEPIANDROSTERONE AND ADIPOSE TISSUE GROWTH 



277 



Control mice were injected with the vehicle 
only. After 20 weeks of treatment the mice 
were killed and tissue processed as de- 
scribed below. 

Experiment 2. Mice were either housed 
individually (50 mice) as described or 
housed five per cage (30 mice) in plastic 
cages wtih bedding. Mice were fed Purina 
mouse chow (5015, 1 1% fat by weight). The 
mice individually housed received pow- 
dered food while the mice housed five per 
cage were given food in a pelleted form. 
One-half of each group were intubated with 
DHEA 450 mg/kg 3x weekly that was 
homogenized in sesame oil. Each control 
group received the vehicle only. Mice were 
treated for 10 weeks and then eight mice 
from each of the four groups were randomly 
selected and were killed and tissue pro- 
cessed as described below. Data presented 
are for the mice that were killed. 

General procedures. At the end of each 
experimental period mice were killed by 
decapitation. Livers were removed for fatty 
acid synthetase activity determination. The 
livers were weighed and a (30%, w/v) ho- 
mogenate was prepared in a 0.15 Af KCl-4 
mA/ MgClz, pH 7.5, solution. The homoge- 
nate was centrifuged at 10,00Q^ for 15 min 
at 4"^ and the supernatant was removed and 
recentrifuged at 100,000^ for 45 min at 4°. 
The high-speed supernatant was stored at 
-70**. Fatty acid synthetase activity was 
determined by the method of Hsu et al, 
(11). Activity was expressed as counts per 
minute per milligram of supernatant pro- 
tein. Protein was determined by the method 
of Lowry £»/ aL (12). 

Both right and left epididymal and ret- 
roperitoneal fat pads were removed. Right 



and left depots were pooled from each site 
and weighed. Two small samples from each 
depot were used to determine fat cell size 
and number by the method of Hirsch and 
Gallian (13). One sample was placed in 2:1 
chloroform: methanol to extract the lipids 
and the lipid was quantitated gravimetri- 
cally (14). The second sample was placed in 
a 2% osmium tetroxide mixture and incu- 
bated for 72 hr at 37°. Fixed cells were 
counted electronically using a Coulter 
Counter (Model ZB). The remainder of the 
adipose tissue was prepared for lipoprotein 
lipase activity determinations. Tissue was 
homogenized (20%, w/v) in 0.25 M su- 
crose- 1 mM EDTA buffer, pH 7.4. The 
homogenate was centrifuged at 11,00Q^ for 
15 min at 4°. The postmitochondrial super- 
natant was removed and frozen at -IQf. 
Lipoprotein lipase activity was determined 
by the method of Schotz et aL (15) as mod- 
ified by Hietanen and Greenwood (16). 

Statistical analysis. Data are presented 
as means ± standard deviations. Results 
from Experiment 1 were analyzed by Stu- 
dent's / test. Results from Experiment 2 
were analyzed first by two-way analysis of 
variance followed by F test for individual 
comparisons (17). Level of significance was 
P = 0.05 or less. 

Results. Experiment I. Body weight and 
food intake of treated and nontreated mice 
are shown in Table I. As can be seen there 
were no differences in body weights of 
treated mice compared to nontreated. In 
addition, food intake was not affected by 
DHEA treatment. Liver weights and liver 
fatty acid synthetase activity are also 
shown in Table I. Neither of these factors 
was affected by DHEA treatment. 



TABLE I. Effects of Intraperitoneal DHEA Treatment" on Body Weight, Food Intake, 
Liver Weight, and Liver Fatty Acid Synthetase Activity in Male C3H Mice (Experiment 1) 



Body weight 
(g) 



Cumulative 

food intake 

(g) 



Liver weight 
(g) 



FAS 
(cpm/mg protein) 



Control 
DHEA 



32.1 ±2.3* 
31.5 ±2.6 



792.3 ± 38.3 
821.0 ±47.6 



1.7477 ±0.2020 
1.7298 ±0.1977 



1.755 ±0.493 
1.570 ±0.350 



" Mice were treated from 6 until 26 weeks of age (25 mg/kg body weight 3 x weekly). 
* Data are means ± standard deviations; N = 6, both groups. 



278 



DEHYDROEPIANDROSTERONE AND ADIPOSE TISSUE GROWTH 



TABLE II. Epididymal and Retroperitoneal Fat Pad Weights and Fat Cell Size and 
Number in Male C3H following Intraperitoneal DHEA Treatment* (Experiment I) 



Epididymal* 



Retroperitoneal* 



Control 



DHEA 



Control 



DHEA 



Pad weight (g) 
Fat cell size 

(/Ag lipid/cell 
Fat cell no./ 

pad(xlO«) 
LPL activity" 



0.9154 ± 0.2460*^ 

0.4915 ±0.1852 

1.3631 ±0.4353 
3.3131 ± 3.7065 



0.7115 ±0.3302 

0.5178 ± 0.2678 

0.9308 ±0.1168* 
2.1784 ± 1.3883 



0.2419 ± 0.0809 

0.6637 ±0.1688 

0.2670 ± 0.0696 
3.2115 ±2.3652 



0.1723 ±0.0804 

0.4637 ± 0.2443 

0.2398 ± 0.0664 
2.5658 ± 1.2319 



" Mice were treated from 6 until 26 weeks of age, 25 mg/kg body weight intraperitoneally injected three times 
weekly. 

* Combined right and left pads. 

*" Data are means ± standard deviations; n = 6, both groups. 

'^ Lipoprotein lipase activity = /imole free fatty acid released x 10* cells. 

* DHEA treated significantly different from respective control groups. 



Fat pad weights are shown in Table II. 
Both epididymal and retroperitoneal fat pad 
weights were reduced by DHEA treatment 
in comparison to nontreated control fat pad 
weights. However, these differences were 
not significant. In the epididymal fat pad, 
DHEA treatment did result in a significant 
decrease in fat cell number (Table II). No 
difference in epididymal fat cell size was 
found. In the retroperitoneal fat pad there 
was no change in fat cell number. There 
was a decrease in retroperitoneal fat cell 
size but this was not significant (Table II). 
Lipoprotein lipase acitivity in both depots 
was decreased by DHEA treatment (Table 
II). However, these differences were not 
significant. 



Experiment 2. Body weight, weight gain, 
and cumulative food intake data are found 
in Table III. These values presented for the 
mice selected for further study were not 
different from the entire group of mice 
used. For example body weights of the 
control one/cage mice were 30.4 ± 2.9 and 
the body weights for the entire 25 mice in 
this group was 30.5 ± 2.4. DHEA adminis- 
tered by intubation resulted in decreased 
body weight in both caging situations 
(Table 111). However, the biggest effect 
from DHEA treatment on body weight and 
weight gain was seen in DHEA five/cage 
mice compared to their control group. In 
addition, significant differences were found 
in the body weights of DHEA five/cage 



TABLE III. Body Weight, Weight Gain, Cumulative Food Intake and Combined Fat Pad 
Weights as a Percentage of Total Body Weight" in DHEA-Treated* Male C3H (Experiment 2) 



Body weight 
(g) 



Initial 



Final 



Weight gain 
(g) 



Cumulative 

food intake 

(g) 



Combined fat pad 

weights, percentage 

total body weight 



Control 1/cage 
DHEA 1/cage 
Control 5/cage 
DHEA 5/cage 



17.8 ± 1.7' 
16.6 ± 1.9 

19.9 ± 1.6 
20.1 ± 1.3 



30.4 ± 2.9* t 
27.1 ± 1.5 
35.1 ± 1.7* 
28.1 ±2.2 



12.6 ± 1.9* 

10.9 ± 1.4 

15.6 ±2.1* 

8.0 ± 1.1 



327.3 ± 36.9* 
343.3 ± 24.4 
225.31: 
205.7 



3.92 ± 0.74*t 
2.91 ± 0.70 
4.61 ± 0.60* 
2.60 ± 0.54 



" Combined epididymal and retroperitoneal fat pads (both right and left) divided by total body weight xlOO. 

** Treated with DHEA 450 mg/kg body weights from 6-16 weeks of age. 

' All data are means ± standard deviation; N = 8 all groups. 

* Control group significantly different from DHEA-treated group. 

t Control 1/cage significantly different from DHEA 5/cage. 

t No statistics done as no individual results were available. 



DEHYDROEPIANDROSTERONE AND ADIPOSE TISSUE GROWTH 



279 



mice compared to control five/cage mice 
from the second week of treatment while 
DHEA one/cage mice body weights were 
significantly different from control one/cage 
at week 1, 4, 9, and 10 (data not shown). 
Food intakes were not different between 
DHEA one/cage and control one/cage mice 
nor between DHEA five/cage and control 
five/cage mice (Table III). Both groups of 
five/cage mice ate less food than one/cage 
mice. The exact explanation for this finding 
is unknown. Food given to the one/cage 
mice was in powdered form and although 
great care was taken to minimize spillage by 
placing food in a small glass cup inside a 
larger cup, it is possible that over 10 weeks 
some unaccountable loss occurred. In ad- 
dition, there may be a higher energy cost 
for mice housed individually as suggested 
by the decreased body weight in one/cage 
mice. 

Liver weights were not different between 
any of the groups (Table IV). However, 
liver weight accounted for a significantly 
larger percentage of total body weight in 
both treated groups (Table IV). No signifi- 
cant differences in liver fatty acid syn- 
thetase activity were found between control 
and DHEA-treated groups (Table IV). 

Epididymal and retroperitoneal fat pad 
weights of both treated groups were signifi- 
cantly decreased compared to their respec- 
tive nontreated groups (Tables V, VI). The 
differences in fat pad weight were much 
greater than those found for body weights. 
For example, DHEA five/cage mice weighed 
15% less than control five/cage, while 
epididymal fat pad weight was decreased 



55%. The effect of DHEA was greater on 
fat pad weights in the five/cage mice than 
in the one/cage mice. In both treated groups 
the combined epididymal and retroperito- 
neal weights accounted for significantly less 
of total body weight than the combined 
weights of fat pads of the untreated groups 
(Table III). 

Cellularity data for epididymal and ret- 
roperitoneal fat pads are shown in Tab